Oecd Economic Surveys: United Kingdom 2003-2004

Special Feature: Product Market Competition and Economic Performance Non-Member Economies Baltic States, February 2000 B...

Author: Organisation for Economic Co-Operation and Development

9 downloads 767 Views 1MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

Special Feature: Product Market Competition and Economic Performance Non-Member Economies Baltic States, February 2000 Brazil, June 2001 Bulgaria, April 1999 Chile, November 2003 Romania, October 2002 Russian Federation, February 2002 Slovenia, May 1997 Federal Republic of Yugoslavia, January 2003

OECD Economic Surveys

United Kingdom

OECD Economic Surveys United Kingdom

Economic Surveys Australia, March 2003 Austria, December 2003 Belgium, February 2003 Canada, September 2003 Czech Republic, April 2003 Denmark, July 2003 Euro area, October 2003 Finland, March 2003 France, July 2003 Germany, January 2003 Greece, July 2002 Hungary, June 2002 Iceland, April 2003 Ireland, July 2003 Italy, August 2003 Japan, January 2003 Korea, March 2003 Luxembourg, September 2003 Mexico, January 2004 Netherlands, January 2002 New Zealand, January 2004 Norway, September 2002 Poland, July 2002 Portugal, February 2003 Slovak Republic, March 2004 Spain, May 2003 Sweden, March 2004 Switzerland, January 2004 Turkey, December 2002 United Kingdom, March 2004 United States, November 2002

Volume 2004/4

Sweden

www.oecd.org

Volume 2004/3 – March 2004

ISBN 92-64-02022-5 10 2004 03 1 P

-:HSTCQE=UWUWWV:

March 2004

ISSN 0376-6438 2004 SUBSCRIPTION (18 ISSUES)

Volume 2004/3 – March 2004

© OECD, 2004. © Software: 1987-1996, Acrobat is a trademark of ADOBE. All rights reserved. OECD grants you the right to use one copy of this Program for your personal use only. Unauthorised reproduction, lending, hiring, transmission or distribution of any data or software is prohibited. You must treat the Program and associated materials and any elements thereof like any other copyrighted material. All requests should be made to: Head of Publications Service, OECD Publications Service, 2, rue André-Pascal, 75775 Paris Cedex 16, France.

OECD ECONOMIC SURVEYS 2004

United Kingdom

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT

ORGANISATION FOR ECONOMIC CO-OPERATION AND DEVELOPMENT Pursuant to Article 1 of the Convention signed in Paris on 14th December 1960, and which came into force on 30th September 1961, the Organisation for Economic Co-operation and Development (OECD) shall promote policies designed: – to achieve the highest sustainable economic growth and employment and a rising standard of living in member countries, while maintaining financial stability, and thus to contribute to the development of the world economy; – to contribute to sound economic expansion in member as well as non-member countries in the process of economic development; and – to contribute to the expansion of world trade on a multilateral, nondiscriminatory basis in accordance with international obligations. The original member countries of the OECD are Austria, Belgium, Canada, Denmark, France, Germany, Greece, Iceland, Ireland, Italy, Luxembourg, the Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, Turkey, the United Kingdom and the United States. The following countries became members subsequently through accession at the dates indicated hereafter: Japan (28th April 1964), Finland (28th January 1969), Australia (7th June 1971), New Zealand (29th May 1973), Mexico (18th May 1994), the Czech Republic (21st December 1995), Hungary (7th May 1996), Poland (22nd November 1996), Korea (12th December 1996) and the Slovak Republic (14th December 2000). The Commission of the European Communities takes part in the work of the OECD (Article 13 of the OECD Convention).

Publié également en français.

© OECD 2004 Permission to reproduce a portion of this work for non-commercial purposes or classroom use should be obtained through the Centre français d’exploitation du droit de copie (CFC), 20, rue des Grands-Augustins, 75006 Paris, France, tel. (33-1) 44 07 47 70, fax (33-1) 46 34 67 19, for every country except the United States. In the United States permission should be obtained through the Copyright Clearance Center, Customer Service, (508)750-8400, 222 Rosewood Drive, Danvers, MA 01923 USA, or CCC Online: www.copyright.com. All other

Table of contents Assessment and recommendations I. Macroeconomic developments, prospects and policy challenges Introduction Recent developments The outlook: growth revives again The main policy challenges Notes Bibliography

II. Reducing the risk of instability from the housing market Introduction The housing market in international context The housing market as a source of macroeconomic instability What is the appropriate policy response? Assessment Notes Bibliography Annex 2.A1. Consumption equation including housing wealth

III. The fiscal challenge: complying with the fiscal rules while raising standards in health and education Introduction The aggregate fiscal position Rising public spending on health care and education Performance management, targets and incentives Long-term public finances Assessment Notes Bibliography Annex 3.A1. An equation for corporation tax receipts Annex 3.A2. Potential output growth estimates

IV. Policies to enhance potential growth Introduction Labour utilisation Closing the productivity gap Assessment Notes

© OECD 2004

9 23 23 23 35 37 41 42 43 43 43 46 51 56 57 60 62 65 65 66 74 82 91 91 94 97 99 100 103 103 103 110 125 127

OECD Economic Surveys: United Kingdom

4

Bibliography Annex 4.A1. Funding of tertiary education in OECD countries

129 133

V. Product market competition and economic performance

141 141 142 150 155 181 185 189

Overview Product market competition and macroeconomic performance Competition legislation and enforcement Regulatory policies Summary and recommendations Notes Bibliography

VI. Some aspects of sustainable development

193 193 198 204 211 213

Climate change Improving living standards in developing countries Sustainable retirement income Notes Bibliography Glossary of acronyms

217

Annex A. Progress on structural reforms

219

Boxes 1.1. 2.1. 3.1. 3.2. 3.3. 3.4. 4.1. 4.2. 4.3. 5.1. 5.2. 5.3. 6.1.

Evaluation of the case for EMU entry The Barker review of housing supply The 2002 and 2003 Budgets and the 2003 Pre-Budget Report The fiscal framework Public Administration Select Committee report on performance targets Activity-based funding, incentives and waiting times in health care New Deal Programmes Upgrading the skills of adults The London Congestion Charge Competition institutions The rail industry – key players and relationships Rolling stock The integration of policies across sustainable development areas

26 55 68 70 86 89 108 116 125 152 178 180 194

Tables 1.1. Recent outcomes and short-term projections 2.1. Household financial liabilities 2.A1.1.Response of consumption to housing wealth 3.1. Public sector finances: selected summary indicators and official projections 3.2. Employment growth in the public sector 3.3. Examples of targets in the Public Service Agreements for health care and education 3.A2.1.Decomposition of trend growth 4.1. New Deal summary statistics 4.2. Educational attainment of the young studying and graduating now 4.3. Continued adult training and education

36 47 63 67 78 83 101 109 114 115

© OECD 2004

Table of contents

4.A1.1.Funding of tertiary education in OECD countries 5.1. Output, employment and productivity 5.2. Hirshman-Herfindahl indices of industry concentration 5.3. Import penetration by manufacturing industry 5.4. Gross domestic expenditure on R&D as a percentage of GDP 5.5. Share of high-technology R&D spending in manufacturing 5.6. Key structural features of the retail distribution sector 5.7. Own-label penetration in European packaged grocery 5.8. Measures of profitability in food retailing 5.9. Regulation indices for professional services 5.10. Prices for unbundled local loop 5.11. Competencies and resources of energy sector regulators 5.12. Electricity market indicators and implementation of the EU Electricity Directive 5.13. Gas market indicators and implementation of the EU Gas Directive 5.14. Electricity and gas retail prices 5.15. Rail infrastructure investment in EU countries 5.16. Rail traffic on the British national rail network 5.17. Percentage of trains arriving on time 6.1. Main indicators: climate change 6.2. Main indicators: trade 6.3. Main indicators: development co-operation 6.4. Producer support equivalents and their components 6.5. Performance indicators: sustainable retirement income

5

136 143 145 147 149 151 159 160 161 163 169 170 171 172 173 176 179 180 195 200 201 202 206

Figures 1.1. 1.2. 1.3. 1.4. 1.5. 1.6. 1.7. 1.8. 1.9. 1.10. 1.11. 1.12. 1.13. 1.14. 1.15. 1.16. 2.1. 2.2. 2.3.

Key indicators in long-term and international perspective Magnitude of recent output gaps Contributions to growth Real and nominal growth differentials between consumption and GDP growth The real exchange rate and terms of trade Real household wealth Consumption, disposable income and mortgage equity withdrawal Relative performance of manufacturing and services External trade The change in monetary and fiscal stance RPIX and HICP inflation The wage share and inflation Real UK per capita GDP compared to other major OECD countries The sources of real income differences Labour utilisation and productivity gaps The contributions of labour utilisation and productivity to trend GDP per capita growth Housing investment and profitability House building House prices relative to personal disposable income, average earnings and rents

© OECD 2004

24 25 27 27 28 28 29 29 30 31 33 34 38 38 39 39 45 45 47

6


2.4. Household interest payments relative to disposable income 2.5. Correlation between real house price growth and consumption 2.6. Effect of an abrupt fall in house prices 3.1. Tax-to-GDP ratio 3.2. Non-North Sea corporation tax receipts 3.3. OECD projections of general government finances 3.4. Health and education expenditure in international perspective 3.5. Speed limits have been reached for public spending 3.6. Earnings and working time in health care and education 3.7. Teacher salaries from an international perspective 3.8. Performance in health care and education 3.9. Waiting times, spending and incentives 4.1. Unemployment and inactivity rates 4.2. Working age claimants of incapacity-related benefits and unemployment 4.3. Inactivity of older workers due to illness or disability 4.4. Human capital explains part of the productivity gap 4.5. Basic literacy of the adult population and of the young 4.6. Educational attainment of the adult population 4.7. Expenditure on tertiary education institutions 4.8. Business investment per worker 4.9. Regulation and investment in information and communication technology 4.10. Pick-up in multi-factor productivity growth and increase in ICT investment 4.11. Government investment 4.A1.1.Tuition fees in tertiary education 4.A1.2.Public loans to students in tertiary education 5.1. Indicators of product market regulation 5.2. Average mark-ups by market structure 5.3. Foreign direct investment outflows and inflows 5.4. R&D expenditure in manufacturing by technology intensity 5.5. Summary indicators of regulation in retail distribution 5.6. Five-firm market concentration in food retailing in EU countries 5.7. New retail floor space in town centres and out of town 5.8. Estimates of incumbent operators’ market share 5.9. Average monthly telephone charges 5.10. Costs of internet access 5.11. Funding and oversight of the rail industry 6.1. Greenhouse gas emissions 6.2. Effect of reforms on public pensions

48 50 51 71 72 73 74 75 77 79 80 90 104 105 106 111 112 113 118 121 122 123 124 134 135 144 146 148 150 157 158 159 167 168 169 178 196 208

© OECD 2004

BASIC STATISTICS OF THE UNITED KINGDOM (2002)

THE LAND Area (1 000 km2): Total Agricultural (2001)

243 186

Major cities (thousand inhabitants, 2001): Greater London Birmingham Leeds Glasgow (local government district)

7 188 976 716 579

THE PEOPLE Population (thousands, mid-2002) Number of inhabitants per km2 Net increase in population, 1991-2001, estimated annual average (thousands)

59 207 244 136

Total labour force (thousands, 2002) Civilian employment (% of total, 2002): Agriculture, forestry and fishing Industry and construction Services

29 934 1.4 24.1 74.5

PRODUCTION Gross domestic product: In £ billion Per head (USD)

1 043.9 26 453

Gross fixed capital investment As a % of GDP Per head (USD)

16.3 4 307

THE GOVERNMENT Public consumption (% of GDP) General government (% of GDP) Current and capital expenditure Current revenue Net debt

20.0 40.5 39.0 31.9

Last general election: 7 June 2001

Composition of House of Commons (number of seats): Labour Conservatives Liberal Other Total

408 163 54 34 659

FOREIGN TRADE Export of goods and services (% of GDP) Main commodity exports (% of total): Chemicals Manufactured goods and articles Electrical machinery Mechanical machinery

26.1

Imports of goods and services (% of GDP)

15.2 23.5 20.7 12.2

Main commodity imports (% of total): Manufactured goods and articles Electrical machinery Road vehicles Mechanical machinery and other transport equipment

29.1 28.2 21.3 12.2 12.4

THE CURRENCY Monetary unit: Pound sterling

December 2003, average of daily rates: £ per USD £ per euro

0.544 0.669

This survey is published on the responsibility of the Economic and Development Review Committee of the OECD, which is charged with the examination of the economic situation of member countries. • The economic situation and policies of the United Kingdom were reviewed by the Committee on 16 December 2003. The draft report was then revised in light of the discussions and given final approval as the agreed report of the whole Committee on 5 January 2004. • The Secretariat’s draft report was prepared by David Turner, Jens Lundsgaard, Maria Maher, Douglas Sutherland, and Michael Wise under the supervision of Peter Hoeller. • The previous Survey of the United Kingdom was issued in January 2002.

Assessment and recommendations Recent macroeconomic performance has been impressive

The UK economy has proved remarkably resilient during the recent downswing with output falling only a little below potential. At the same time, inflation has remained close to the target and the unemployment rate is among the lowest in the OECD. More recently the economy has been gaining considerable momentum, well ahead of the euro area. This strong performance is underpinned by wide-ranging structural reforms and sound macroeconomic policy frameworks. A pro-active monetary policy successfully stabilised activity and inflation, helped by a strong transmission channel operating through the housing market. The government budget balance was in substantial surplus at the peak of the cycle and this has enabled fiscal policy to be strongly supportive of growth during the downswing, although the government deficit has now become sizeable. In the context of this impressive macroeconomic performance the decision that was taken in June 2003 to wait further before holding a referendum on entry into the European monetary union is perhaps unsurprising. All in all, the United Kingdom seems well placed to take advantage of the global recovery and move towards a more broadly based growth that relies less on consumption and housing wealth.

The main challenges are to contain instability in the housing market, to spend public money efficiently and to close the productivity gap

Looking ahead, three major policy challenges arise:

© OECD 2004

– In the short term, it will be important to gradually withdraw the policy stimulus to avoid running into supply constraints at a premature stage of the ongoing upswing. Early monetary tightening will also contribute to reducing the risk of macroeconomic instability that may emerge from the housing market, but there is also an issue as to what other policy measures might contribute to this.

10


– In the medium term, improving the quality of public services in a cost effective way in such priority areas as health and education will be a major challenge. This may be helped by slowing down the ongoing build-up in public expenditure. A smooth phasing in of new spending could also ease the difficult trade-off between raising taxes or departing from the government’s “golden rule” and Code for Fiscal Stability. – In the long term, the challenge is to raise growth performance further, by reducing the still sizeable “productivity gap” with the best performing OECD countries, while inactivity of some groups is still of concern. The housing market is a source of resilience but also of potential instability

The strong overall performance has been driven by a buoyant services sector and domestic demand, while net exports and the manufacturing sector have been a drag on activity, reflecting both the sharp appreciation of the exchange rate between 1996 and 1998 and the weakening of activity in continental Europe in recent years. In sharp contrast to the euro area, consumption, which has been boosted by terms of trade gains and more recently by wealth effects associated with sharply rising house prices, remained vigorous throughout. Indeed, private consumption has been the prime driver of growth. Yet, the propensity of house prices to surge in response to lower interest rates, which contributed to the resilience of the UK economy, may also become a source of fragility in the context of an inelastic housing supply and rapid build-up in household debt, increasing the risk of setbacks in the ongoing recovery. The OECD projects growth of 2¾ and 3 per cent in 2004 and 2005, respectively, underpinned by a recovery in foreign trade and the further rapid rise in public spending. Despite the assumption of a gradual decline in house price inflation, private consumption is likely to remain strong. Given the increased importance of mortgage equity withdrawal in driving consumption, a potential risk is that house prices surge ahead again in the short run instead of slowing down, but then fall abruptly which could trigger a sharp reduction in consumption growth. However, the situation is different from the late 1980s because of low household income gearing with high housing wealth in relation to income. In addition, there is now greater scope for monetary policy to react, particularly because inflation currently remains under control.

© OECD 2004

Assessment and recommendations

11

Monetary policy tightening should be pursued

Despite this increasing risk to macroeconomic stability, but in line with the current consensus among central banks, the Monetary Policy Committee of the Bank of England has repeatedly rejected the idea that monetary policy should explicitly respond to changes in asset prices, except insofar as they affect future inflationary pressures. Nevertheless, they raised the repo rate for the first time in nearly four years in November 2003 on the grounds that inflation is likely to pick up in the context of a vigorous recovery. The labour market is already tight with the unemployment rate near its structural rate and the wage share at a high level which in the past has signalled a pick-up in inflation. Gradually raising interest rates further would reduce the risk of a larger and more abrupt tightening later on, and hence reduce the risk of triggering instability in the housing market and the wider economy.

Measures to deal with housing should focus on the supply side

Beyond an early tightening of monetary policy, policy measures should focus on the supply side of the housing market. Reforms of the planning system which improve the supply of housing would be welcome, not only because a more elastic supply would tend to moderate swings in house prices, but also because increased housing supply and associated lower prices would be highly desirable from a wider economic and social perspective. The planning issue has implications for the efficiency of the housing and retail sectors and a variety of services sectors as well as capital accumulation, which in international comparison suffers from a low rate of investment in structures. The government is doing more, including through new legislation, to ensure that local authorities meet objectives for house building. However, given the inevitable lags before there is an appreciable effect of such policies on the housing stock, other measures acting on the demand side should also be considered, not least because reducing swings in house prices could itself be an important factor in encouraging housing supply. On the fiscal side the most promising option is to reform the “council tax” to relate it more closely to current valuations of property. The government is due to revalue properties for the purposes of the council tax by 2007, but revaluations should then be updated more regularly. At the same time it should be made less regressive. More regular

© OECD 2004

12


updating would also imply that rising real house prices would generate higher taxes and so tend to automatically damp the stimulus to consumption and reduce the risk of macroeconomic instability. The advent of statutory regulation of mortgage markets by the Financial Services Authority in 2004 is also to be welcomed. There is a need, in particular, to ensure that prudent lending requirements are not relaxed during a boom in house prices. The “golden rule” may come under strain

The government conducts fiscal policy to meet both its “golden rule”, that over the course of the cycle it should not borrow to finance current expenditure, and its “sustainable investment rule”, that public sector net debt remains below 40 per cent of GDP. The 2003 Pre-Budget Report projects a significantly weaker current balance from now on compared to the 2003 Budget projections, while the deficit could breach the Maastricht Treaty’s deficit limit in 2003. On the one hand, the golden rule is expected to be met over this cycle, projected to end in 2005-06, with the current balance returning to surplus at the start of the next cycle. On the other hand, the more pessimistic OECD projections suggest that, although the golden rule would just be met in the current cycle, a sizeable structural deficit would persist. In this case, meeting the fiscal rules in the future would imply revenue or spending adjustments. It would be advisable to start implementing these changes during the current upswing rather than postpone measures and thereby risk reinforcing the downswing of the next cycle. At the same time firm cost control is also advisable. Ensuring a safe margin against the risk of a persistent structural budget deficit may reduce the need for future increases in interest rates.

Public expenditure growth should be slowed on efficiency grounds…

For many years outlays in health and education were low in international comparison, and raising spending was required to achieve ambitious outcome targets. Real expenditure on health is currently planned to rise at an annual rate of 7¼ per cent until 2007 and on education by 5¾ per cent until 2005. Although there are signs of better service outcomes and improved quality of services, some of these represent a continuation of earlier trends that were visible before the substantial recent spending increases were initiated, and the speed with which the spending increase is

© OECD 2004


13

taking place is generating cost pressures within the public sector. A more gradual increase in health and education spending could therefore improve the chance of locking in improved performance rather than higher costs. This would also enable the government to carefully pilot and implement various promising innovations – such as Foundation Hospitals, which give hospitals more operational flexibility, activity-based funding mechanisms and Fast Track Surgeries. As an additional benefit, extending the period over which the increase takes place would yield savings to public finances in the short term. All in all, assessing the effectiveness of such a large build-up in public health and education spending will always be fraught with difficulties. The current signs of overheating and overstretching may be more benign than they look. But the “option value” of waiting to ensure best value for money also seems high. … and longerterm there is a strategic choice between higher taxes and user charges

© OECD 2004

The deteriorating fiscal position also highlights a basic strategic choice in improving health care to a level common in most of continental Europe as well as further raising standards in education: either to accept the corresponding level of taxes and mandatory contributions that are typical in continental Europe, or to expand private funding in ways that do not compromise equity concerns. To avoid large tax rises, users could contribute part of the funding of improving health care and education. An obvious case for this is in higher education, because universities are publicly funded and the private returns are probably the highest in the OECD due to a wide earnings differential between persons with and without university education. Letting graduates pay a larger share of the study costs would be both fair and economically efficient, and the government’s plan to introduce a graduate contribution scheme is both innovative and welcome. It resolves the credit constraints facing students from poor backgrounds by giving a loan to fund increased tuition fees which will have to be repaid after graduation unless the person’s income falls below a minimum threshold. Expanding higher education based on income-contingent graduate contributions, while maintaining the large publicly funded improvements made in early childhood and compulsory education, is the most direct way to achieve equity in access to higher education and education

14


outcomes more generally. In health care, private funding on a large scale is excluded by the government’s principle that access to health care should not depend on ability to pay. However, within this framework there is scope for extending cost-sharing. For example, additional user charges to cover the extra cost of non-generic drugs or charging for missed appointments, can improve incentives, while raising some revenue. Getting incentives right and finding the appropriate level of spending is all the more important in the United Kingdom because the most powerful fiscal pressures as a result of population ageing over coming decades will come from health care rather than pensions. Performance targets in the public sector must be supported by funding incentives

To get maximum value for the extra spending on public services, continuous improvements in the performance management framework are necessary. The United Kingdom is at the forefront of applying quantitative outcome focussed performance targets in the public sector. These targets have helped drive improvements in public services, created transparency and raised ambitions for service delivery. The framework now needs to evolve further – with fewer targets, which are limited to the main outcomes and key government priorities, with clearer focus on outcomes and with improved metrics. However, targets do not guarantee value for money; they only affect behaviour if service providers have an incentive and feel motivated to meet the targets. Activity-based funding for hospitals, for example, can help reduce waiting times and lengths of stay, but, within the framework of the recently agreed contract, incentive-pay for hospital doctors should be introduced faster and more broadly than currently planned.

Long-term growth can be raised by increasing activity rates

The UK’s trend growth has been boosted over the last two decades by the substantial decrease in structural unemployment. However, there has been virtually no increase in the activity rate. Indeed, the male activity rate has shown a consistent downward trend and has been accompanied by a similar rise in men reporting long-term sickness or disability as the main reason for inactivity. There has also been a trebling of persons on invalidity benefit during the 1980s and the first half of the 1990s, and since 1995 the number claiming incapacity benefit has continued to rise, albeit at a

© OECD 2004


15

slower rate, to a level which is now two and a half times the unemployment claimant count. This may partly be because the rest of the income support system has been tightened so much that incapacity benefits have become a pressure valve. “New Deal” active labour market policies to deal with particular disadvantaged groups, notably lone parents, have been successful in raising their employment rates. The voluntary New Deal for Disabled People (NDDP) is the major employment programme available to all those claiming incapacity benefits. The new “Pathways to work” scheme offers improved financial incentives and support. Together these represent steps in the right direction, although an expansion in the scale of these programmes will probably be needed to make significant progress. A productivity gap with the best performing countries persists

The United Kingdom has achieved a commendable rate of potential growth, which is above the European average. As a consequence, the GDP per capita gap with the major continental European countries has nearly closed. However, the gap with the best performing OECD countries – such as the United States, Canada, and Australia – has hardly diminished and stems mainly from weaker levels of productivity. Closing the “productivity gap” with major competitors has been an important theme motivating structural reforms and a multitude of policy initiatives address this weakness. The most important among these are: a “Skills Strategy” to raise skill levels throughout the workforce; increased government infrastructure investment; and a recent overhaul of competition policy.

Raising skills is a promising route to reduce the productivity gap

The recent OECD Growth Project highlighted the importance of human capital in explaining differences in growth performance among OECD countries. The United Kingdom compares poorly with its major competitors on both basic literacy and vocational skills, suggesting that this has been a major factor holding back productivity growth. It may explain why there has been no boost to multi-factor productivity growth recently, despite a substantial increase in ICT investment. The education system is now improving, but too many people still leave education early and relatively few get an apprenticeship, skilled craft or technician level qualification. Improving the educational opportunities of this

© OECD 2004

16


group should remain a policy priority along with providing more resources to universities via the graduate contribution scheme. Training of persons with limited skills already in work is valuable from a personal as well as productivity perspective, but only if improvements are sufficient to warrant the costs of tuition and the loss of production from being out of work. In this context, cost-sharing should be developed to ensure that relevant course content is chosen. Increased government investment in transport is welcome but congestion charges may also be needed

The government is committed to raising general government investment, which over decades has been sacrificed as a short-term palliative for achieving budget cuts, but which under the current fiscal rules should be insulated from such pressures. However, even plans to double the share of public net investment in GDP to just over 2 per cent will still leave government investment below the level in most other major OECD countries. Probably the area where insufficient infrastructure investment has most impinged on long-term growth prospects is transport. The government is committed to a substantial increase in investment in transport, especially on roads. However, this is unlikely to stem traffic congestion, which is already much worse and more costly to business than in most other countries. Following the early signs of success of the London congestion charge, the government should consider introducing road charging schemes on an experimental basis, with a view to deciding whether road use charges, which better reflect congestion costs, could complement and partially replace fuel taxes.

The recent overhaul of legislation should help to further promote competition

Competitive pressures appear to be relatively strong in the United Kingdom, with economic and administrative regulations inhibiting competition and barriers to trade amongst the lowest in the OECD. Nevertheless, there is scope for improvement and the recent overhaul of competition legislation should lead to progress in this regard. Active enforcement, together with market investigations which help to identify competition problems in markets that are not due to infringements of the competition law, will help to ensure that markets are competitive. But the government will need to continue to limit the use of “public interest” intervention and ensure the enforcement agencies’ inde-

© OECD 2004


17

pendence. More emphasis should be placed on enforcement of horizontal price fixing and the Office of Fair Trading (OFT) has already started investigations. To underpin the credibility of the enforcement programme, the OFT should apply the stronger sanctions that are now available against horizontal price fixing. In the longer term, attention needs to be paid to potential overlap between the OFT and the Competition Commission (CC). As the OFT’s range and capacity expand the extent of the CC’s distinct contribution may need to be reconsidered. Planning restrictions and market power are problems in the retail sector

© OECD 2004

In the retail sector, the recent Competition Commission enquiry concluded that in general the supermarket sector operates in a competitive environment and that price differentials with the European Union could mainly be explained by the strong exchange rate. However, market power, both buying and selling power on the part of large retailers, remains a problem. In particular, the enquiry found that retailers have used their market (buyer) power to the detriment of suppliers and competitors. In light of this, the OFT negotiated a code of practice to govern buyer-supplier relationships in this sector, although it is not clear that this will resolve the problems and the competition authorities will need to remain vigilant. As regards entry into the sector, the government’s recent approach to planning has made new large scale entry very difficult. Competition in the market is thus impeded both by inhibiting entry and by preventing firms from growing in size to achieve their full productivity potential. Over the 1990s, productivity growth of the UK’s retail sector has been particularly disappointing as compared to that of the United States. While the productivity gains arising from large scale entry are more difficult to realise in the United Kingdom due to land scarcity and overcrowding, the government is currently re-examining the appropriateness of the current planning restrictions. They could be relaxed without compromising social objectives. These restrictions have been fairly successful in achieving social objectives such as preserving town centres but come at a cost in terms of competition in the sector, thereby harming consumer interests.

18


Much progress has been made in professional services but the government should reconsider the case for opening entry into pharmacies

There have been many reforms in the professional services sector in recent years. An important step is that the regulations of self-regulatory bodies are no longer exempt from competition legislation and professional bodies have undertaken a number of actions towards removing or easing restrictions that the OFT identified as inhibiting competition. In its response to the findings of a recent OFT enquiry into pharmacies calling for open entry in the sector, the government, citing concerns over the current shortage of pharmacists, has left in place entry restrictions that inhibit the development of competition. However, the OFT report found that consumers are currently well served geographically and that, in general, there was no shortage of pharmacies. The government should reconsider the recommendations of the OFT and allow open entry into pharmacies, subject only to demonstrated professional qualifications. The Government will review this area in 2006.

Industry regulators need to remain vigilant in electricity, gas and telecoms

The opening up of the electricity, gas and telecoms sectors has led to increased productivity, but international comparisons suggest that there is still scope for prices to fall, particularly in telecoms. Reforms in electricity and gas have led to benefits for consumers where prices are generally below the EU average, but household electricity prices remain comparatively high. Prices for mobile and business telephone charges also remain comparatively high with business telephone charges amongst the highest in the European Union, probably due to market power on the part of the incumbent. In all of these cases there is a need for the relevant industry regulator to remain active to prevent abuses of dominant position and ensure consumers reap the full benefits from liberalisation.

Another reorganisation of the rail sector solves some problems but raises new issues

Continuing problems in the rail sector need to be resolved. Recent actions, through tighter specification of outputs and guidance about planning and investment, show that the government is taking the lead in resolving some of these problems. The Strategic Rail Authority (SRA) is playing a stronger role in specifying franchises and a temporary role in assuring the operation of trains pending the award of competitive franchises. While these recent steps overcome the most serious weaknesses of the privatised rail system, problems

© OECD 2004


19

regarding incentives and responsibilities remain to be resolved. In particular, there is still a need for an independent regulator, and the government should clearly define the lines of responsibility and reinforce and ensure the regulator’s independence. The SRA should consider granting longer term franchises, as the current policy of short franchises with tightly specified performance criteria is unlikely to provide incentives for necessary long term investments. Ensuring income adequacy for the retired will remain a challenge

© OECD 2004

In the long run, there are also challenges to ensure that economic development is sustainable in both social and environmental terms. Unlike many other OECD countries, the United Kingdom’s public pension system is financially sustainable, as the basic state pension will be uprated by prices not by earnings. Thus, the relative importance of the Basic State Pension will decline over time, partially offset by the contribution from the (earnings related) State Second Pension, which will increase over time. In addition, public provision is being increasingly targeted at poor pensioners. The authorities have set income adequacy on retirement as an important goal in recent reforms. These reforms should ensure that pensioners are guaranteed a basic income, if take up of the increasingly important means-tested benefit, the Pension Credit, is high. The Pension Credit rewards those with savings or income from employment within limits and from age 65, and in this respect is an improvement on the Minimum Income Guarantee. Nevertheless, the operation of the system will introduce disincentives for some individuals to save and work at the margin. The Government believes it is the responsibility of individuals to choose how and how much to save for retirement and whether or not to continue working beyond the age of eligibility. In this light, if income adequacy on retirement remains a problem, the authorities may need to consider raising the level of the basic state pension, although such decisions would have to be made in the light of their effect on public finances. Proposed reforms to toughen the regulatory environment for occupational pensions and simplify the complex pension system are welcome. While further adjustments to the pension system may be necessary, it is important to establish stability in pension system rules and allow time for individuals to digest the implications of the latest wave of reforms.

20


Climate change policy should ensure that targets are met as cost effectively as possible

Meeting the Kyoto Protocol related target should not present any major difficulties to the United Kingdom, given the wide range of measures included in the climate change strategy. However, the efficiency of the policy package could be improved by ensuring that marginal abatement costs are as uniform as possible across the different policies and measures. Participation in the EU’s emission trading scheme offers an opportunity to reformulate the strategy towards this end. While the proposed EU scheme could usefully build on the strengths of the UK’s already existing emission trading scheme, the UK’s participation in the EU’s emission trading scheme would help to create more targeted abatement incentives, in particular in the emission-intensive power generation sector.

Summing up

The performance of the UK economy has been impressive in recent years, underpinned by wide-ranging structural reforms and sound macroeconomic policy frameworks. The OECD projects growth above potential in 2004 and 2005, with unemployment remaining low, but instability stemming from the housing market is a risk. Monetary policy should continue to tighten to reduce future inflationary pressures and the risk of a further sharp rise in house prices from becoming a potential source of fragility. At the same time, the current rigidity of house supply should be eased by reforming land planning, while a reform of the regressive council tax could help to reduce the risk of housing market instability. Over the projection period, government spending will continue to rise as a share of GDP, reflecting the government’s ambitious targets in priority areas. Slowing spending growth may be warranted on efficiency grounds: bottlenecks have become apparent, while time is needed to pilot new innovative approaches. More generally, getting incentives right is essential to meet the ambitious service targets. Slowing spending growth would also benefit the public finances. An income-contingent contribution scheme for university graduates is an equitable way forward; other forms of user charges not compromising equity concerns could improve efficiency. Both are preferable to raising taxes. Trend growth has already been boosted by the sharp decline in structural unemployment, but measures are still needed to reduce the number of inactive, especially those

© OECD 2004


21

on incapacity benefit. While educational performance of students has risen, policy needs to further address the lack of vocational skills as one element in a strategy to reduce the productivity gap with the best performers elsewhere. In this respect, competition is also important. The government has strengthened the competition framework and active enforcement along with the market studies of the OFT and the CC’s market investigations should help to ensure that markets become more competitive. However, planning restrictions and market power are a problem in the retail sector and industry regulators will need to remain vigilant in the electricity, gas and telecom sectors. Ensuring macroeconomic stability, while addressing the remaining weaknesses through further structural reforms offers the prospect of continuing strong economic performance.

© OECD 2004

I.

Macroeconomic developments, prospects and policy challenges

Introduction Since the mid-1990s, GDP growth, inflation and unemployment have been remarkably stable in the United Kingdom (Figure 1.1). Nowhere else in the OECD has economic activity remained so consistently close to trend over this period (Figure 1.2).1 The United Kingdom has also been amongthe most resilient economies during the recent downturn. Apart from Canada, it is the only G7 economy, for which output has not fallen by significantly more than one percentage point below potential. Moreover, it has been one of the few European countries (large or small) to display such a high degree of robustness during the downswing (Figure 1.2).2 At the same time the unemployment rate has remained continuously close to 5 per cent. In 2003, it will be the lowest among the major seven economies, while deviations of inflation from the target have been as small as could have been reasonably expected. This impressive macroeconomic performance is a testament to the strength of the current macroeconomic policy framework as well as more than two decades of structural reform.3 It has also been an important factor behind the decision to postpone a referendum on entry into the Economic and Monetary Union (EMU) (Box 1.1). In this generally favourable context, the United Kingdom seems well placed to take advantage of the ongoing world recovery and to move towards a more broadly based growth that relies less on consumption and housing wealth, but extends to so far lagging exports, investment and manufacturing activity. Recent and prospective macroeconomic developments and policy are discussed below. They provide the background for the assessment of two major policy challenges in the future, relating to the housing market and fiscal policy. These policy challenges, together with those of a more structural nature, are highlighted at the end of this chapter. Recent developments Growth has been led by private consumption since the mid – 1990s with net exports acting as a drag on activity in every year since 1996 (Figure 1.3).

© OECD 2004


24

Figure 1.1. Key indicators in long-term and international perspective1

United Kingdom

Euro area

OECD

% change

8

% of potential GDP

A. Real GDP

8

B. Output gap

6

6

4

4

2

2

0 0

-2

-2

-4

-4

-6

-6 1980

1985

1990

1995

2000

1980

1985

1990

1995

2000

% of labour force

-8 % change

20

D. Private consumption deflator 2

C. Unemployment rate 12

15 9 10 6 5

3 0 1980

1985

1990

1995

2000

1980

1985

1990

1995

2000

% of GDP

6

0 % of GDP

E. Government net lending

4

F. Current account

4 2

2 0

0

-2 -2

-4 -6

-4

-8 -10 1980

1985

1990

1995

2000

1980

1985

1990

1995

2000

-6

1. OECD projections for 2003. 2. OECD excludes high inflation countries. Source: OECD.

© OECD 2004


25

Figure 1.2. Magnitude of recent output gaps As a percentage of potential GDP 5

5 Average absolute output gap, 1995-2003 Largest annual output gap since 2000 1

4

4

3

3

2

2

1

1

0

0 (-2.9)

-1

UK

AUS DNK

ITA AUT

CAN NZL

BEL CHE

EURO SWE NLD PRT FRA ISL IRE USA DEU NOR JPN ESP GRC FIN

-1

1. A positive output gap denotes output be below potential output. Source: OECD.

However, in nominal terms the excess of consumption over output growth appears unremarkable compared with many G7 countries (Figure 1.4). This underlines the importance of relative price changes following the rise of the real effective exchange rate by 25 per cent between 1996 and 1998 (Figure 1.5). Although the household saving rate fell by around 3 percentage points between 1997 and 1998, between then and early 2002 the increase in consumption was largely underpinned by growth in disposable income. More recently, however, personal disposable income has slowed, partly due to tax increases, and private consumption has outpaced personal disposable income since early 2002. The housing market has become an increasingly important factor in two respects: the rise in housing wealth has almost entirely offset the effect of the fall in equity prices on household wealth since 2000 (Figure 1.6); and mortgage equity withdrawal was running at close to a record high of 6 per cent of disposable income in the first half of 2003 (Figure 1.7). As house prices have risen on average by nearly 15 per cent per annum since 1999, the vulnerability of consumption and the wider economy to an abrupt fall in the housing market has increased. The appreciation of the effective exchange rate by over 25 per cent between 1996 and 1998 has also been the principle cause of stagnation in the manufacturing sector, compounding the long – run decline relative to GDP, which is common to all developed economies (Figure 1.8). The appreciation also led to sharp market share losses. The increase in the volume of exports of goods and

© OECD 2004


26

Box 1.1.

Evaluation of the case for EMU entry

The government announced its decision to postpone any referendum on EMU entry in June 2003 concluding that in the absence of sustainable and durable convergence or sufficient flexibility to cope with potential difficulties, UK membership was not in the national interest at present. The decision is based on an evaluation of five economic tests of which only one is judged to have been met (H.M. Treasury, 2003): 1) Are business cycles and economic structures compatible so that the UK and others could live comfortably with euro interest rates on a permanent basis? Not met. The UK cycle is found to exhibit a greater degree of cyclical convergence with the euro area countries compared to the last assessment in 1997, and the EC Treaty convergence criteria for inflation, long-term interest rates and government deficits and debt are all met. However, structural differences with the euro area remain, of which the housing market is highlighted, so the test is not met. 2) If problems emerge is there sufficient flexibility to deal with them? Not met. The UK labour market was judged to have become more flexible since 1997 as reflected in a lower level of unemployment not resulting in wage pressures, less regional variation in unemployment rates and fewer long-term unemployed. Progress has also been made in reforming labour, capital and product markets in both the United Kingdom and the euro area. Nevertheless, this flexibility was judged to be insufficient to cope with the likelihood of increased inflation volatility within EMU, so the test was judged not met. 3) Would joining EMU create better conditions for firms making long-term decisions to invest in Britain? Unclear. EMU membership would likely increase investment via more foreign direct investment, lower transaction costs in a larger financial market and possibly lower long-term interest rates, but this is judged to be largely contingent on the first convergence criterion being met. 4) What impact would entry into EMU have on the competitive position of the UK’s financial services industry, particularly the City’s wholesale markets? Met. EMU membership is expected to benefit the UK’s financial service industry. 5) In summary, will joining EMU promote higher growth, stability and a lasting increase in jobs? Not met. Improved prospects for growth, employment and stability would occur, but only once sustainable and durable convergence had been achieved and so the test was judged not met in the light of the above, and especially in the outcome of the first test. The Treasury will provide its update on whether conditions have changed when releasing the budget in April 2004, but it would seem unlikely that the situation will change much in just one year given the existing cyclical divergence between the United Kingdom and the euro area and the need to establish a “clear and unambiguous economic case for membership”.

© OECD 2004


27

Figure 1.3. Contributions to growth Per cent

Per cent

7

7 Private consumption Public consumption Investment

6

Stockbuilding Net exports

6

5

5

4

4

GDP growth

3

3

2

2

1

1

0

0

-1

-1

-2

1995

1996

1997

1998

1999

2000

2001

2002

2003 1

-2

1. OECD projections. Source: Office for National Statistics and OECD.

Figure 1.4.

Real and nominal growth differentials between consumption and GDP growth1 Annual average percentage points, 1995-2002

1.2

1.2

1.0

Real

1.0

Nominal

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0

0.0

-0.2

-0.2

-0.4

-0.4 Canada

France

Germany

Japan

Italy

United States

United Kingdom

1. The bars show the difference between the annual average growth in consumption and GDP over the period 1995-2000 in both real and nominal terms. Source: OECD.

© OECD 2004


28

Figure 1.5.

The real exchange rate and terms of trade

135

135 Real effective exchange rate 1 Terms of trade 2

130

130

125

125

120

120

115

115

110

110

105

105

100

100

95

95

90

1995

1996

1997

1998

1999

2000

2001

2002

2003

3

90

1. Measured with respect to relative consumer prices. 2. Goods and services, national accounts definition. 3. OECD projections for 2003. Source: OECD.

Figure 1.6. Real household wealth1 £ billion, 2000 prices, end year 4500

4500 Housing Equity Other

4000

4000

3500

3500

3000

3000

2500

2500

2000

2000

1500

1500

1000

1000

500

500

0

1988

1990

1992

1994

1996

1998

2000

2002

0

1. Housing wealth is calculated net of loans secured on dwellings. Source: Bank of England.

© OECD 2004


29

Figure 1.7. Consumption, disposable income and mortgage equity withdrawal Per cent

Per cent

8

8

6

6

4

4

2

2

1

0

-2

0

Nominal consumption 1 Personal disposable income 2 Mortgage equity withdrawal

1995

1996

1997

1998

1999

2000

2001

2002

2003

3

-2

1. Year-on-year percentage change. 2. In a per cent of personal disposable income, figures for 2003 are based on 4-quarter sum to 2003 Q2. 3. OECD projections for 2003. Source: OECD.

Figure 1.8. Relative performance of manufacturing and services 1995 = 100

135

135 A. United Kingdom

130

135 B. United States

130

125

Services

125

120

120

115

115

110

110

130

105

105

100

100

1996 1998 2000 2002

Source: OECD and Eurostat.

© OECD 2004

95

125

Services Services

120 115 110

Manufacturing

Manufacturing

95

C. Euro area

Industry (excluding construction)

105 100

1996 1998 2000 2002

1996 1998 2000 2002

95


30

services was 13 percentage points less than the growth of export markets between 1996 and 2003 and there has been a substantial deterioration in the trade balance on goods, although this has been partly offset by a growing surplus on investment income and continued surpluses of non-factor services (Figure 1.9). The current account deficit is expected to be about 2½ per cent of GDP in 2003. This is, however, not yet at a level where it is likely to become a source of exchange rate instability, particularly since part of this deficit is due to the more advanced stage of the recovery relative to the other main European countries. Further evidence of sectoral imbalances is provided by the persistent differential between services and goods inflation.4 Although this differential is generally positive in all countries, reflecting differences in productivity growth, in the United Kingdom services inflation was 5 to 6 percentage points per annum ahead of goods inflation during 2002, whereas in the other major European countries the differential was only on average about 2 to 3 percentage points. However, the differential in the UK has begun to narrow in recent quarters. Figure 1.9. External trade As a per cent of GDP

4

4 Current balance

2

2

0

0

-2

-2

-4

-4 Trade balance Non-factor services, net

-6 1995

1996

1997

Factor income, net Transfers, net

1998

1999

-6 2000

2001

2002

2003 1

1. OECD projections for 2003. Source: Office for National Statistics.

Real business investment grew on average by 9½ per cent per annum over the second half of the 1990s, after not growing at all during the first half. This led to capital deepening with some consequent improvement in productivity growth (see Annex 3.A2 to Chapter III) as well as some catch-up in the level of investment per worker relative to other major competitors (see Chapter IV). How-

© OECD 2004


31

ever, the changes are again less dramatic in nominal terms, with the share of nominal business investment in GDP only rising by one percentage point over the second half of the 1990s to a level which was still slightly below that in the other G7 countries. The share of business investment in GDP has since fallen back and the level of real business investment was little changed in mid-2003 from 2000. Given the resilience of the economy over this period, the recent stagnation of investment may partly reflect companies diverting funds into company pension schemes following the fall in equity markets, but heightened global uncertainty has been an additional factor. The fiscal stance A more supportive fiscal policy has been a major reason why the United Kingdom has weathered the recent downturn better than most other OECD countries, particularly those in the euro area (Figure 1.10), with the cyclically-adjusted balance, as conventionally measured, decreasing by around 3½ per cent of GDP between 2000 and 2003. This large swing is mainly due to rapid spending increases in priority areas and unexpected revenue shortfalls partly related to the fall in equity markets, while discretionary tax increases in 2003 have reduced the expansionary impulse. The relative stability of output means that the cyclical downturn has played a relatively small role in the widening of the actual deficit. Figure 1.10.

The change in monetary and fiscal stance Since 2000

1

1

Fiscal expansion 7

Fiscal expansion 7

2005 +

6

+ 2003

5 4

x

2005

2

x

1 0

3 United States

2 + 2001

Euro area

0x

00

0 0

4

2003

x 2002

United Kingdom

5

+ 2002

x 2004

3

-1

6

+ 2004

1

2

1

2005 0 2004

0 3

4

2

5

6

Monetary easing (percentage points)

1. A positive fiscal shift denotes an increase in the cyclically-adjusted deficit as a share of GDP. 2. Monetary easing denotes a decline in the nominal short-term interest rate. Source: OECD.

© OECD 2004

-1


32

The overall deficit (on a Maastricht basis) is estimated by the OECD to amount to 3 per cent of GDP in 2003, a bit above the euro area average, but significantly below that of Germany, France and the United States. The United Kingdom could afford this sizeable fiscal stimulus because going into the downturn the fiscal position was strong: net lending was higher, and government net debt lower, as a proportion of GDP than for any other G7 country in 2000. With public expenditure planned to outpace GDP growth over the coming years, and in the absence of announced discretionary tax changes, the fiscal stance, according to OECD projections, is expected to be mildly expansionary with annual increases in the noncyclical deficit averaging 0.35 per cent of GDP between 2003 and 2005. This implies that the government deficit (on a Maastricht basis) would remain close to 3 per cent even as the output gap is closed. However, as discussed further in Chapter III, there are significant difference between OECD and Treasury fiscal projections, with the latter implying a steady fall in the government deficit to below 2 per cent of GDP by fiscal year 2007 as well as a modest tightening (rather than loosening) in the fiscal stance going forward as a consequence of greater buoyancy in tax revenues and, to a much lesser extent, the easing of one off spending pressures due to the UK international commitments affecting fiscal years 2002 and 2003. A different concern is that the deflator for public consumption has recently been increasing at an annual rate of between 7 and 8 per cent, although it is not immediately clear how far this reflects capacity constraints in the public sector as opposed to difficulties in measurement. Sir Tony Atkinson has been asked to undertake a review of the future development of measures of government output, productivity and associated price indices so as to advance methodologies. The issue of capacity constraints as well as measuring public sector output is considered further in Chapter III. Monetary policy The monetary stance The Monetary Policy Committee (MPC) of the Bank of England increased the repo rate by ¼ percentage point in November 2003, reflecting growing concerns about future inflationary pressures. This was the first rise in nearly four years. At the onset of the downturn a series of cuts brought the repo rate down from 6.0 per cent in early 2000 to 4.0 per cent by the end of 2001. The rate was unchanged during 2002, but cut again by 25 basis points in both February and July 2003 to 3.5 per cent – the lowest level in 40 years. Judged by the proximity of RPIX inflation to the 2½ per cent target, the operation of monetary policy has remained impressive (Figure 1.11). While up until recently there has been a tendency to undershoot the target, consistent with a bias to over-predict inflation, a

© OECD 2004


33

Figure 1.11. RPIX and HICP inflation Year-on-year percentage change

4.0

4.0

3.5

3.5

3.0

3.0 Target

2.5

2.5

2.0

2.0

1.5

1.5

1.0 0.5 0.0

1.0

RPIX 1 Harmonised consumer price index

1995

1996

1997

1998

0.5

1999

2000

2001

2002

2003

0.0

1. Retail price index, all items excluding mortgage interest payments. Source: Office for National Statistics.

recent independent assessment concluded that the bias was “as small as one could reasonably expect” (Pagan, 2003). While there are no obvious signs of a widespread pick-up in price inflation, there has been a rise in inflation expectations based on survey evidence in late 2003. Growth in money and credit aggregates has also picked up ahead of the growth of nominal GDP, with narrow definitions of money (M0) increasing by nearly 8 per cent and broader measures (M4) by 7½ per cent in the year to the third quarter of 2003. The most immediate source for future inflationary pressure comes from the labour market with unemployment stable at about 5 per cent, which is somewhat below the OECD estimate of the structural rate. While there has yet to be any sign of any generalised pick-up in wage inflation, public sector wages have increased more strongly than in the private sector, with most of the recent employment growth being accounted for by the public sector. The wage share including the public sector has increased to levels that are well above historical norms (Figure 1.12). Recent research suggests that there is a stable relationship between inflation and the wage share over the last 30 years, supplementary to the effect of the output gap, as embodied in a “new Phillips curve” (Batini et al., 2000).5 The only previous periods in the last 30 years when this measure of the wage share has reached a similar level – the mid-1970s, the late 1970s and the early 1990s – were all periods when inflation picked up strongly.6 Given the pick-up

© OECD 2004


34

Figure 1.12. The wage share and inflation Per cent 82

Per cent 25 20

80

15

78 Inflation (right scale)

76

10 5

74 1

0

Wage share (left scale)

72

-5

70

-10

68 1960

1965

1970

1975

1980

1985

1990

1995

2000

-15

1. The wage share is adjusted to include the self-employed as per Batini et al. (2002) and includes general government. Inflation is measured by the change in the GDP deflator. Values for 2003 are based on data to 2003Q2. Source: OECD.

in growth and the prospective closing of the output gap, little or no slack in the labour market and acceleration in money and credit aggregates, monetary policy should continue to tighten, albeit gradually. Additional cost pressures may also emerge from a delayed response to the exchange rate depreciation since the beginning of 2003 and higher social insurance contributions implemented at the time of the 2003 Budget. The new inflation target The Chancellor announced in the Pre-Budget Report that the measure of inflation that is targeted has switched – effective from 10 December 2003 – from the RPIX, retail prices excluding mortgage interest payments, to the Consumer Prices Index (CPI) – the name by which the National Statistician has decided that the Harmonised Index of Consumer Prices (HICP), the European Union’s standardised measure of inflation, will be known in the United Kingdom. The National Statistician has also said that this should not be interpreted as implying that there is any intention to develop the CPI differently from the HICP. The CPI and the HICP will remain one and the same index. The new name will be simpler, emphasise the CPI’s role as a UK index and be consistent with the naming of other price indices. The level for the new CPI inflation target is being set at 2 per cent and the switch to the new target applies from 10 December. The new target is ½ percentage point below the RPIX-based target, reflecting differences in the way

© OECD 2004


35

that RPIX and CPI inflation are measured.7 The level of the new target is set to be consistent with the old target in two years’ time, the typical forecast horizon for monetary policy. The gap between the two inflation measures in the middle of the year was far apart, reflecting the large contribution from housing depreciation costs to the former measure while owner-occupied housing costs are entirely excluded from the latter, but has narrowed over recent months and in October 2003 stood at 1.3 percentage points. This could lead to some short-term issues in communicating changes in monetary policy, given that RPIX inflation is above the current target but CPI inflation is below the likely new target. It should not, however, represent any fundamental policy shift as monetary policy is based on expected inflation and, providing house price inflation moderates, the gap between the two measures of inflation should decline. Indeed, in the past the MPC has been willing to raise interest rates when inflation has been below the target, if it expected that it would rise over the future. In principle, and except for the fact that a suitable housing cost component for this index has not yet been developed, CPI is a superior measure of inflation. The switch also makes sense as a means of easing any possible future transition to EMU. The outlook: growth revives again Following a surprisingly weak first quarter of 2003, which may have been dragged down by confidence effects associated with the Iraq conflict, GDP rebounded and is likely to grow by close to 2 per cent for the year as a whole, while output in the euro area is likely to grow by only ½ per cent. Private consumption, together with rising public expenditure, have been the main drivers of activity in 2003. The path the economy follows thereafter will be partly dependent on how the housing market develops. A benign scenario would see a gradual fall in house price inflation over the course of the next few years, which would allow a gradual decline in the contribution of private consumption to growth without any abrupt adjustment – this is the assumption underlying the central scenario.8 Exports should recover with world trade, and may make some modest gains in market share following the depreciation of sterling since the beginning of 2003, although net exports are likely to remain a drag on activity given a weaker recovery in the rest of Europe. Companies have made good progress in tackling balance sheets and have moved to a position of net lending in the last four quarters after several years of borrowing. However, capital gearing remains relatively high and so companies may be reluctant to borrow further to finance investment. In addition, companies are likely to continue diverting funds into reducing pension fund deficits. These two factors may restrain growth in business investment during the recovery phase.

© OECD 2004


36

Following the stock market decline, the accumulated pension fund deficit for private sector non-financial corporations is estimated to be £160 billion, about 15 per cent of GDP and additional increases in contributions to pension funds of the order of ¾ to 1½ per cent of GDP have been estimated may be needed over coming years (Confederation of British Industry, 2003).9 However, under alternative assumptions and using more recent equity and bond valuations, the deficit could be considerably lower. The need for annual provisioning will be reduced if stock markets rise, but in the coming years provisioning will largely be decided on the basis of recent and current depressed valuations.

Table 1.1. Recent outcomes and short-term projections Percentage change, constant prices 2000

2001

2002

20032

20042

20052

Private consumption

4.6

3.1

3.6

2.4

2.4

2.2

Government consumption

1.9

1.7

2.4

3.4

1.7

2.4

Gross fixed investiment Private consumption Residential Business investment Government investment

3.6 3.4 – 0.5 4.8 6.3

3.6 2.9 0.9 3.6 12.0

1.8 1.3 16.1 – 3.5 7.4

2.9 1.3 1.0 1.5 19.3

4.9 3.3 4.9 2.7 19.6

6.4 4.4 4.3 4.5 21.6

Final domestic demand Stockbuilding1

3.9 – 0.1

2.9 – 0.2

3.0 – 0.2

2.7 – 0.3

2.7 0.3

2.9 0.2

Total domestic demand

3.8

2.7

2.9

2.4

3.0

3.1

9.4 9.1 – 0.1

2.5 4.5 – 0.6

– 0.9 3.6 – 1.4

– 0.9 1.1 – 0.6

6.5 7.0 – 0.4

8.0 8.0 – 0.4

GDP at constant prices

3.8

2.1

1.7

1.9

2.7

2.9

GDP deflator GDP at current prices

1.4 5.2

2.3 4.5

3.2 5.0

2.8 4.8

2.2 4.9

2.5 5.5

Memorandum items Consumer price index, harmonised Underlying price index, harmonised Retail price index (RPIX)

0.8 0.1 2.1

1.2 1.1 2.1

1.3 1.5 2.2

1.3 1.2 2.8

1.7 1.6 2.6

2.3 2.3 2.7

Total employment Unemployment rate

1.1 5.5

0.8 5.1

0.7 5.2

0.9 5.0

0.5 4.9

0.6 4.8

– 2.1 3.9

– 1.8 0.7

– 1.8 – 1.5

– 2.7 – 2.9

– 3.5 – 2.9

– 3.6 – 3.2

2.5 1.2

2.9 0.5

2.7 – 0.5

2.5 – 1.1

2.5 – 0.9

2.4 – 0.4

Exports of goods and services Imports of goods and services Net exports of goods and services1

Current account balance, % of GDP Government net lending, % of GDP Potential output, whole economy Output gap, whole economy

1. Contribution to changes in real GDP (as a percentage of real GDP in the previous period). 2. OECD projections are tose presented in the OECD Economic Outlook No. 74, December 2003. Source: OECD.

© OECD 2004


37

According to this scenario, growth would approach 2¾ and 3 per cent in 2004 and 2005, respectively. The composition of growth would become gradually more balanced, while the output gap would close by the end of 2005 (Table 1.1). This should also be consistent with RPIX inflation close to the 2½ per cent target and CPI inflation close to the new 2 per cent target. The main policy challenges Despite this impressive macroeconomic performance and promising growth prospects highlighted above, concerns remain about some potential imbalances in the economy and how they will unwind in the future. There is first a risk that the housing market boom unfolds more abruptly than expected with adverse consequences for consumption and economic stability. The policy challenge of reducing the risk of instability stemming from the housing market is considered in Chapter II. There are also fiscal policy challenges ahead. The deficit has increased, while spending is likely to continue to outpace output growth. Without corrective action, on OECD projections the deficit is likely to remain close to 3 per cent of GDP in coming years, although as discussed further in Chapter III, recent official projections imply the deficit will fall to 2 per cent of GDP by fiscal year 2007 largely because of greater buoyancy in tax revenues. The major challenge for fiscal policy will be to deliver on the ambitious outcome targets for public spending in priority areas in a cost effective way, while respecting the existing fiscal rules. These challenges are considered in Chapter III. A recent study concluded that evidence of the benefits of two decades of structural reform in the United Kingdom was provided by the halting of “the nearly century – long trend in relative economic decline of the United Kingdom relative to its historic competitors France and Germany” (Card et al., 2002). Indeed, over the last decade the gap in GDP per capita with the major continental European countries has been substantially closed (Figure 1.13, left panel). It might still be argued that the United Kingdom has much to learn from these countries as they achieve a similar level of GDP per capita with substantially lower labour input because the level of productivity is correspondingly higher (Figure 1.14). This inference may, however, be misleading in the light of the strong inverse correlation between productivity levels and labour utilisation across OECD countries (Figure 1.15). In a number of countries high productivity reflects the fact that lowskilled jobs have been priced out of the market, due to a combination of tax and benefit systems, employment protection and collective bargaining arrangements, with a corresponding chronic lack of employment opportunities for low-skilled workers. In this light, a more challenging peer comparator group for the United Kingdom comprises the major English-speaking countries, Australia, Canada and

© OECD 2004


38

Figure 1.13. Real UK per capita GDP compared to other major OECD countries In 1995 prices and 1995 PPPs UK GDP per capita = 100

150

France

UK GDP per capita = 100 Germany

1

Italy

Australia

Canada

United States

150

140

140

130

130

120

120

110

110

100

100 1975

1980

1990

1995

2002

1975

1980

1990

1995

2002

1. For Germany the data for 1990 refer to 1991. Prior to 1991, data for Germany are based on figures for West Germany only but have been scaled down on the basis of relative per capita GDP in 1991 to avoid a discontinuity. Source: OECD.

Figure 1.14. The sources of real income differences 2002, 1995 PPPs Percentage gap with respect to UK GDP per capita =

Effect of 1,2 labour resource utilisation

+ Effect of 1,3 labour productivity

United States Canada Australia Germany France Italy

0

10

20

30

40

50

-30

-20

-10

0

10

20

-10

0

10

20

30

40

1. Percentage gap with respect to the United Kingdom level. 2. Labour resource utilisation is measured as trend total number of hours worked divided by population. 3. Labour productivity is measured as trend GDP per hour worked. Source: OECD.

© OECD 2004


39

Figure 1.15. Labour utilisation and productivity gaps Percentage gap with respect to the United States level, 2002, 1995 PPPs Labour resource utilisation gap with US 1

10

10 Japan

5

5

correlation coefficient = -0.9

0

Australia

Canada

0

-5

-5 United Kingdom

-10

-10

-15

-15 European Union

-20

-20

-25

Germany

-25

France

-30

-30 Italy

-35 -40 -30

-25

-20

-15

-10

-5

0

-35 5

10

-40

Labour productivity gap with US 2

1. Labour resource utilisation gap is measured as trend total number of hours worked divided by population. 2. Labour productivity gap is measured as trend GDP per hour worked. Source: OECD.

Figure 1.16.

The contributions of labour utilisation and productivity to trend GDP per capita growth Per cent per annum, average 1995-2002 Labour resource utilisation 1 +

Contributions to growth =

labour productivity 2

Australia United Kingdom United States Canada Germany France Italy

-1

0

1

2

3

-1

0

1

2

3

-1

0

1

1. Labour resource utilisation is measured as trend total number of hours worked divided by population. 2. Labour productivity is measured as trend GDP per hour worked. Source: OECD.

© OECD 2004

2

3

40


especially the United States. This group of countries has both significantly higher labour productivity and labour utilisation10 than the United Kingdom (Figure 1.14) and has since the mid – 1990s, unlike the other major continental European countries, managed to sustain productivity growth while increasing labour utilisation (Figure 1.16). Moreover, the United Kingdom has failed to significantly reduce the GDP per capita gap with this group of countries over the last decade (Figure 1.13, right panel). The major structural policy challenge is therefore to close the GDP per capita gap with this group of better performing OECD countries. This will most clearly require a closing of the productivity gap, but would be further helped by continued increases in labour utilisation, issues which are further addressed from a policy perspective in Chapter IV.

© OECD 2004


41

Notes

1. More formally, between 1995 and 2003 the average absolute deviation of output from estimated potential for the United Kingdom is the lowest in the OECD according to OECD estimates. 2. The only OECD European countries where output has fallen less below potential than for the United Kingdom since 2000 are Spain, Sweden and Ireland. 3. In addition to sound policies the Governor of the Bank of England has also attributed the “nice” (non-inflationary consistently expansionary) nineties to a slice of “lady luck”, in that “economic surprises alternated between good one year and bad the next, rather than adding up to one damn thing after another”, (King, 2003). 4. As measured by the harmonised index of consumer prices (HICP). 5. Batini et al. (2000) estimate a “new Phillips curve” obtained from a structural dynamic model of price-setting extended to capture employment adjustment costs and the openness of the UK economy. Two different measures of the wage share are modelled, one with and the other without the public sector. Over the sample the authors consider that there is little to choose between the two on empirical grounds, although the authors comment: “It might be argued that the concepts of labour and capital shares only really make sense with regard to the market sector of the economy.” On the other hand rapid expansion of public sector employment and wages might be expected to have inflationary spill over effects on the private sector in the context of a tight labour market. 6. Alternative explanations of the rise in the wage share, such as an increase in competition in the goods market or a lower cost of capital, would not necessarily imply pressures from the labour market and impending inflationary pressures. 7. Differences between the RPIX and HICP inflation rate reflect differences in coverage of the two indices and differences in methodology used to construct them. The HICP uses geometric rather than arithmetic averages to combine highly disaggregated prices and this in itself reduces inflation fairly consistently by about ½ per cent per annum relative to the RPIX measures. The most important difference in terms of coverage is that HICP does not include owner – occupier housing costs, whereas RPIX does. Since housing costs, at least as measured in the RPIX, have consistently risen by more than other prices this reduces HICP inflation by around a further ¼ of a percentage point relative to RPIX inflation. For further details see the box on page 38-39 of the Bank of England’s May 2003 Inflation Report. 8. This is consistent with the assumption which has been habitually made by the MPC in making their projections in recent years, although it has been repeatedly confounded.

© OECD 2004

42


9. Prudential rules require companies to make financial provisions to eliminate pension fund deficits over no longer than the effective service life of the individual fund, and for UK corporate pension schemes the average remaining service life is 10-12 years. 10. Note that labour utilisation here refers to hours. The UK employment rate in 2002 was a similar level to that in the United States and Canada and above that in Australia.

Bibliography Batini, N., B. Jackson and S. Nickell (2000), “Inflation Dynamics and the Labour Share in the UK”, External MPC Members’ Discussion Papers, No. 2, November, available from www.bankofengland.co.uk/. Card, D. and R. Freeman (2002), “What Have Two Decades of British Economic Reform Delivered”, National Bureau of Economic Research Working Paper, No. 8801, February, available from http://papers.nber.org/papers/W8801. Confederation of British Industry (2003), “Focus on Investment: The Impact of Pension Deficits”, Economic Brief, July, available from: www.cbi.org.uk/home.htm. H.M. Treasury (2003), “UK Membership of the Single Currency: An Assessment of the Five Economic Tests”, Cm 5776, June, available from www.hm-treasury.gov.uk/. King, M. (2003), Speech given at the East Midlands Development Agency/Bank of England Dinner, Leicester, 14 October, text available from www.bankofengland.co.uk/speeches. Pagan, A. (2003), “Modelling and Forecasting at the Bank of England”, available from www.bankofengland.co.uk/.

© OECD 2004

II.

Reducing the risk of instability from the housing market

Introduction Over the last couple of years the housing market has been a major consideration in the setting of monetary policy as demonstrated by numerous references in the minutes of the Monetary Policy Committee (MPC). The high interest rate sensitivity of the housing market strengthens considerably the transmission channel of monetary policy, in marked contrast with most continental European countries. On the other hand, there are a number of episodes over the last three decades when developments in the housing market have either been a cause of, or exacerbated, macroeconomic instability. The challenge is therefore to retain the inherent strength of the current framework, i.e. a strong reactivity to interest rate changes, while reducing its drawbacks in terms of potential instability and possible misallocation of resources in the long run. The government is committed to a comprehensive programme to improve the functioning of the housing market, addressing both supply and demand issues. Such measures are seen as encouraging greater convergence with the euro area economies, but are also considered “beneficial in their own right to improve the stability and flexibility of the United Kingdom housing market and wider economy”. The remainder of this chapter puts the UK housing market in international context, evaluates it as a source of macroeconomic instability, and finally considers what the appropriate policy response is, both in the current cycle and over the longer term. The housing market in international context The share of the housing stock that is owner-occupied is 70 per cent in the United Kingdom, slightly higher than the EU average, but much higher than in France (55 per cent) or Germany (43 per cent) (Ball, 2003). The share that is rented from private landlords has remained roughly stable at about 10 per cent over the last two decades, following a substantial post-war decline. This levelling off partly reflects the removal of rent controls as well as the gradual phasing out of the most blatant tax advantages in favour of owner-occupation, such as tax relief on mortgage interest payments. The share of the housing stock rented from local authori-

© OECD 2004

44


ties has more than halved over the last two decades to 14 per cent, mainly as a result of “right to buy” discounted sales to sitting tenants as well as sales to social housing institutions. The latter are mainly non-profit making housing associations and trusts that now account for about 7 per cent of the housing stock.1 The UK mortgage market is amongst the most liberalised in the OECD and there is a high degree of competition among mortgage lenders.2 The availability of 100 per cent loans places the United Kingdom at the least restrictive end of the wide range of loan-to-value ratios that are applied across OECD countries (Girouard and Blondal, 2001).3 Transaction costs associated with buying a house are also relatively low: comparisons for the mid-1990s suggest that, for an average – priced property, transactions costs were about 2 per cent of the purchase price in the United Kingdom, compared to over 7 per cent in Germany and Italy and nearly 14 per cent in France.4 The high degree of liberalisation of the mortgage market, as well as the relatively high level of housing transactions (associated with low transactions costs), is also reflected in the scale of “mortgage equity withdrawal”,5 as households use the increased value of their property as collateral to borrow at rates of interest that are substantially below those on unsecured loans. At the end of 2002 a typical consumer taking out a loan secured on property would face an interest rate of 5½ per cent compared to 10½ per cent on an unsecured bank loan or 15½ per cent using a credit card.6 In contrast, in many countries where deregulation of the mortgage market has been less extensive, the household sector has been permanently injecting equity into housing (Girouard and Blondal, 2001). The high level of mortgage debt and high share of variable rate mortgages also means that the monetary transmission mechanism acting through the housing market is relatively strong in the United Kingdom, certainly in comparison to the major continental European countries, as emphasised in the Treasury’s EMU study (H.M. Treasury, 2003a). The United Kingdom also stands out in international comparison in another respect. Estimates of the elasticity of housing supply with respect to real house prices are typically much lower than for other OECD countries (Swank et al., 2002 and Malpezzi and Maclennan, 2001) and are particularly low for the South of England (Meen, 1996). There has been some recent pick-up in real housing investment in response to the increased profitability of housing investment, proxied as the differential between indices of house purchase prices and the housing investment deflator (Figure 2.1). Nevertheless, despite the increase in house prices since the mid-1990s, housing investment has averaged less than 4 per cent of GDP, lower than in most other OECD countries and little changed compared to the depressed housing market years of the early 1990s. In terms of houses completed the supply response is even more muted with the number of dwellings completed falling to a post-war low in 2001, but recovering slightly in 2002 (Figure 2.2). The fall in the late 1990s in part reflects the decline in council house building by local authorities. At the regional level the extent of the supply constraint is even more

© OECD 2004


45

Figure 2.1. Housing investment and profitability Index, sample average = 100

160

160

150

Profitability of investment

1

150

Real housing investment

140

140

130

130

120

120

110

110

100

100

90

90

80

80

70

1970

1975

1980

1985

1990

1995

2000

70

1. The profitability of housing investment is measured as the ratio of residential property prices to the implicit residential investment deflator. A 4-quarter moving average of both series is taken to smooth high frequency volatility. Source: Office for National Statistics, Office of the Deputy Prime Minister and OECD.

Figure 2.2. House building Permanent dwellings completed, by tenure Thousand

Thousand

400

400 Private Registered social landlords Local authorities

350

350

300

300

250

250

200

200

150

150

100

100

50

50

0

1970

1975

1980

Source: Office of the Deputy Prime Minister.

© OECD 2004

1985

1990

1995

2000

0

46


apparent with new house building falling in the South-East where price pressures have been most acute. To a large extent this reflects land constraints and the high population density of the United Kingdom as well as the regional concentration of economic activity. However, as discussed further below, there is also evidence that planning restrictions may be hampering supply. The housing market as a source of macroeconomic instability Inelastic housing supply has meant that the trend increase in real house prices (relative to the private consumption deflator) has been among the strongest in the OECD (Girouard and Blondal, 2001). Periods of rising house prices have sometimes gathered momentum and led to speculative bubbles, a feature which may be more likely in a liberalised mortgage market. This occurred most notably in the early 1970s and late 1980s: between 1971 and 1973 real house prices rose 46 per cent and then fell 36 per cent over the following four years; between 1986 and 1989 real house prices rose by 50 per cent and fell by 31 per cent over the following four years. Some empirical modelling of UK house prices has captured such speculative effects through the inclusion of a non-linear term in the predicted rate of return applied as a cubic specification of the previous period’s capital appreciation.7 More recently, real house prices have risen 50 per cent in the five years to 2002, and nominal house price inflation in mid-2003 was running at an annualised rate of around 15 per cent,8 having repeatedly exceeded the MPC’s predictions of an imminent slowdown.9 In relation to average earnings and personal disposable income the current level of house prices has, respectively, exceeded and is close to the peak that occurred at the height of the late 1980s housing boom prior to the subsequent sharp fall (Figure 2.3). This has led a number of commentators to conclude that house prices are over-valued and a sharp correction is due although others cite structural reasons why the house price to earnings ratio may have increased.10 In relation to rents, at least as measured by the rent component of the retail prices index, house prices are also close to their late 1980s peak, although this may provide a less reliable benchmark as much of the rental sector, particularly up until the 1990s was regulated. Rents on private “assured” tenancies (for which rents are not regulated), data for which are only available over a shorter sample, have been rising less quickly than house prices: in the three years to 2002 the average annual rate of increase of such rents was about 6 per cent whereas the average rate of increase of house prices was roughly double that.11 There are a number of channels by which instability in the housing market could either lead to general macroeconomic instability, or else could magnify the consequences of another adverse shock. One set of concerns focuses on the increased indebtedness of households and the consequences of coping with higher debt service if interest rates rise or there is some other adverse shock to

© OECD 2004


47

Figure 2.3. House prices relative to personal disposable income, average earnings and rents1 1970-2002 = 100

160

160 Relative to per capita disposable income Relative to rents Relative to average earnings

140

140

120

120

100

100

80

80

60

1970

1975

1980

1985

1990

1995

60

2000

1. House prices are those provided by the Office of the Deputy Prime Minister; rents are the rent component of the retail prices index. Source: Office of the Deputy Prime Minister, Office for National Statistics and OECD.

demand.12 Between 1999 and 2002 the increase in household financial liabilities in the United Kingdom, both as a percentage of disposable income and in relation to financial assets, has been the most rapid of any G7 country over this or any other 3- year period since 1990 (Table 2.1). The level of financial liabilities is now among the highest in the G7, second only to Japan in relation to disposable income and to Germany in relation to financial assets. The rise in household liabilities has been heavily influenced by developments in the housing market, with long-term loans secured on dwellings making up nearly three-quarters of household financial Table 2.1. Household financial liabilities Per cent of income

Japan United Kingdom Germany Canada United States France Italy Source:

OECD.

© OECD 2004

Per cent of financial assets

2002 level

Change 1999-2002

2002 level

Change 1999-2002

136.7 129.2 112.0 111.7 108.8 76.0 35.3

11.6 17.4 –2.2 –0.1 4.9 0.7 1.5

28.3 34.6 41.4 32.1 25.0 21.9 12.3

0.0 13.1 0.6 0.9 5.3 0.9 2.0


48

liabilities. While the strong rise in house prices over recent years has led to increased borrowing for house purchase, it has also significantly boosted the value of housing wealth. The increase in household liabilities as a percentage of total household assets has been far more muted at only 3½ per cent. The rise in debt service payments following the collapse of the housing market and rising interest rates at the end of the 1980s led to a sharp increase in the savings ratio and a substantial increase in mortgage arrears and repossessions. An important difference, however, between the current situation and that prevailing in the late 1980s is that interest rates and inflation are much lower. The low level of nominal interest rates means that despite the record level of indebtedness, household interest payments (dominated by mortgage interest payments) remain a relatively low proportion of personal disposable income, and well below the level prevailing at the beginning of the 1990s (Figure 2.4). Even so, if regular payments of mortgage principal are included, there has been a steady rise in a broader measure of debt service payments in relation to disposable income since the mid-1990s. But for this wider measure of debt service to rise to a similar proportion of household disposable income as experienced at the beginning of the 1990s, interest rates would have to rise by 8 to 9 percentage points.13 This would seem highly unlikely given the current low inflation. Consistent with the relatively low level of interest repayments, there are few signs yet of

Figure 2.4. Household interest payments relative to disposable income1 Per cent of post-tax income

16

16

15

15

Including regular repayments of mortgage principal

14

14

13

13

12

12

11

11

10

10

9

9

8

8

7 6

7

Interest payments only

1988

1990

1992

1994

1996

1998

2000

2002

6

1. Dashed lines indicate averages over period shown. Source: Bank of England estimates; Office for National Statistics.

© OECD 2004


49

widespread financial distress in the household sector in terms of repossessions or mortgage arrears, both of which remain close to historical lows. While households appear much less exposed to an adverse shock than in the late 1980s, aggregate data may disguise the extent to which individual households are exposed. Information from the British Household Panel Survey (BHPS) for 2000, the latest available year, suggests that, although the households with the highest absolute levels of debt tended also to have the highest incomes and net wealth, the youngest and lowest – income households increased their debt-toincome ratios by most – and from the highest levels – between 1995 and 2000 (Cox et al., 2002). These are also the households that are most vulnerable to financial and other shocks likely to increase financial stress, such as unemployment or increases in interest rates.14 However, according to a more recent survey conducted by NMG Research for the Bank of England15 on holdings of unsecured debt, the share of households reporting debt to be a burden seems little changed in recent years. In the BHPS surveys from 1995 to 2001, around 10 per cent reported their debts to be a heavy burden and 29 per cent somewhat of a burden, similar to the NMG survey. The rapid growth of unsecured debt in recent years has not, therefore, as yet led to any overall increase in the degree of financial distress reported by households, although the survey does point to a small group of heavily indebted individuals who continue to face substantial problems in servicing their debt. A further reason for concern is that recent rates of house price inflation are unsustainable, and an abrupt change could have a large and rapid effect on consumers’ spending, as evidenced by the high correlation between short – run changes in consumption and real house prices (Figure 2.5). Indeed, statistically annual changes in consumers’ expenditure are better correlated with contemporaneous changes in house prices than with changes in real personal disposable income.16 The rapidity of this response probably reflects the importance of the transmission mechanism through mortgage equity withdrawal. Econometric analysis by both the Treasury and the Bank of England, as well as similar work reported in Annex 2.A1, confirms that changes in housing wealth have a relatively large short – run impact on consumption behaviour which “over-shoots” the long-run effect.17 An implication of the large short-run effects of house wealth on consumption is that a sharp slowdown in the growth rate of house prices, and especially an abrupt fall, would have a large effect on consumption. For example, the consumption equation reported in Annex 2.A1 suggests that from the current situation where house prices have been running at 15 to 20 per cent per annum, an immediate levelling off in nominal house prices (i.e. zero house price inflation) would lead to a fall in the consumption-income ratio by about 2 percentage points over four quarters. The full impact on GDP of this decline in consumption would however be muted both because of a partial offset as imports fall and due to the likely policy response. To explore this option more fully a scenario of falling house prices is

© OECD 2004


50

Figure 2.5.

Correlation between real house price growth and consumption 1970-2002

Annual change in real consumption 8

Annual change in real consumption 8

6

6

4

4

2

2

0

0

-2 -4 -20

-2

correlation coefficient = 0.83

-15

-10

-5

0 5 10 15 20 Annual change in real house prices1

25

30

-4

1. Real house prices are measured as the house price series from the Office of the Deputy Prime Minister deflated by the private consumption deflator. Source: Office of the Deputy Prime Minister and OECD.

examined using the OECD’s Interlink model as a variant to the central projection described in Chapter I.18 A scenario with falling house prices The central projection considered in the previous chapter was based on a gradual decline in house price inflation over the coming years. However, historical experience suggests that following periods of exceptional growth, asset price adjustments are rarely smooth.19 An alternative is considered here, whereby house price inflation rebounds to rates of around 20 per cent until end-2004 at which point there is an abrupt fall in the absolute level of house prices by 20 per cent.20 Reflecting the large short-term impact of housing wealth there is a rapid effect on consumption with the savings ratio rising by around 4 percentage points in the space of a year. In the absence of any policy response GDP growth would be lowered by around 1 percentage point in 2005, with much of the negative impact of the shock to consumption offset by lower imports. If it is assumed that shortterm interest rates are cut by 100 basis points over the course of the year following the start of the fall in house prices, but are restored to the same level as in the base scenario by the end of 2006, then GDP growth would still be reduced by about ½ a percentage point in 2005 (Figure 2.6). The general point illustrated by the scenario is that an abrupt fall in the level of house prices, particularly if

© OECD 2004


Figure 2.6.

Effect of an abrupt fall in house prices

Baseline scenario

30

51

Simulation

A. House prices

B. GDP growth

Year-on-year percentage change

Change from previous period, annual rate

4.0

20

3.5

10

3.0

0

2.5

-10

2.0

-20

1.5

-30

6 5

2002 2003 2004 2005 2006 2007 2008

2002 2003 2004 2005 2006 2007 2008

C. Private consumption growth

D. Savings ratio

Change from previous period, annual rate

As a per cent of disposable income

4

1.0

9 8 7

3

6

2

5

1 4

0

3

-1 -2

2

-3

1

-4

2002 2003 2004 2005 2006 2007 2008

2002 2003 2004 2005 2006 2007 2008

0

Source: Simulations of Interlink model (see text for details).

immediately preceded by a period of high house price inflation, would be likely to have substantial effects on the real economy, and it is doubtful that monetary policy reactions would be able to impact quickly enough to offset them. What is the appropriate policy response? In line with the current consensus among central banks, the MPC has repeatedly rejected the idea that monetary policy should explicitly respond to changes in asset prices, except insofar as they affect future inflationary pressures.21 The arguments against a more pro-active policy towards asset prices include the difficulty of assigning a single instrument to more than one target, the difficulty in

© OECD 2004

52


identifying what the equilibrium level of asset prices is (or equivalently whether a bubble has developed) as well as the unpredictable response of asset prices to monetary policy. There is now a compelling case for raising interest rates further on the grounds of impending inflationary pressures, as argued in Chapter I, and the case for doing so is further strengthened to the extent that an early and more gradual tightening would reduce the risks of an abrupt correction in the housing market. This still, however, raises the issue of whether there are either alternative policy instruments or reforms that might be better used to target the housing market with a view to reducing the risk of it becoming a serious source of macroeconomic instability. Should fiscal instruments be used to counter instability in the housing market? The manner in which housing is taxed can influence the variability of housing prices. Van den Noord (2003) provides clear evidence that among euro area countries those with more generous income tax treatment of mortgage repayments also tend to have greater variability in house prices. However, tax relief on mortgage interest payments has already been phased out in the United Kingdom. Another possibility is to vary stamp duty as a discretionary instrument to damp housing market fluctuations by changing the rates in relation to the house price cycle.22 Recent experience from Ireland suggests that changes in stamp duty can indeed have a marked influence on house prices (OECD, 2003), as highlighted by the speed with which a recent increase was subsequently reversed. There is also evidence for the United Kingdom that a temporary stamp duty “holiday” in 1992 reversed a downward trend in property transactions (H.M. Treasury, 2003b). However, using a transactions tax to address macroeconomic instability may have a cost in terms of reduced labour mobility. Cross-country evidence suggests that there is an inverse relationship between residential mobility and the level of transactions costs associated with buying a property, and recent empirical analysis for the Netherlands suggests that transaction costs have a very strong negative effect on the owners probability of moving: a 1 percentage point increase in transactions costs reduces residential mobility rates by 8 per cent (Van Ommeren and Van Leuvensteijn, 2003).23 The sale of a principal private residence is exempt from capital gains tax in the United Kingdom as in most other OECD countries. However, some countries, for example Germany, charge capital gains tax if the property is sold within a certain period of the initial acquisition. While such a tax might be expected to reduce the variability of house prices it would also tend to have an adverse effect on mobility. In general, taxes on property values have an advantage over transactionsbased taxes, such as stamp duty and capital gains tax, because they do not penalise mobility as they are neutral with respect to the length of ownership. The

© OECD 2004


53

“council tax” on property values raised around £16 billion (1½ per cent of GDP) in financial year 2002, around one-quarter of all local authority revenue, being paid on the capital value of property (with some exemptions). It is, however, highly regressive, not increasing in proportion to property values: the absolute level of council tax payments on properties valued at £30 000, £300 000 and £3 million varies in ratio of 1 to 2½ to 3.24 The tax is also currently based on 1991 housing valuations and although there are plans for work to begin in 2005 on revaluing properties, this will not be complete until 2007. Updating these valuations more regularly, say every two to three years, is now technically feasible.25 In these circumstances, introducing a constant rate or even progressive structure might contribute to reducing macroeconomic instability caused by the housing market. Rising house prices would then generate higher taxes and if homeowners anticipate this, the upswing may be automatically damped. Given the historical experience of repeated episodes where real house prices have risen by the order of 50 per cent over a period of a few years, it is clear that such a tax could have a large effect on consumption behaviour.26 Muellbauer (2003a, 2003b) argues that property taxes have contributed to stability of the housing market and private consumption in Denmark and that they might serve as a model for reform of the council tax in the United Kingdom. In Denmark there is a flat-rate tax of 1 per cent on the annually assessed market value of owner-occupied housing up to a threshold at which a higher rate is paid.27 For the purposes of the property tax the value of dwellings is assessed every second year, and in the year in-between these assessment values are adjusted according to the general index for the price of dwellings.28 The Danish economy has indeed been very stable over recent years without any major swings in house prices despite having a relatively liberalised housing market. Simulations of a multi-country macroeconometric model suggest that following a shock to consumption the operation of automatic stabilisers in Denmark are the strongest for any EU country.29 Moreover, the house price equation in the Bank of Denmark’s macroeconometric model suggests that a 1 percentage point increase in the property tax would reduce house prices by around 7 per cent over four quarters acting solely through the user – cost of housing, which would be additional to any effect the tax change had on disposable income (Bank of Denmark, 2003). Do mortgage markets need reforming? The UK mortgage market, as previously noted, is more liberalised and competitive as compared to most OECD countries. Given the benefits to consumers as well as the wider economy, particularly through promoting labour mobility from easy access to mortgage credit, there should be a reluctance to impose restrictions. Nevertheless, there is a concern that during the period of a boom, criteria for lending may become overly relaxed. For example lending at higher multi-

© OECD 2004

54


ples of income – over 3 times single income or 2¾ times joint income – roughly doubled over the four years to 2002 to over 40 per cent of new mortgage loans (FSA, 2003). Whilst lending at higher income multiples exposes borrowers to greater risks, these risks need to be considered in the context of lower and more stable interest rates than in previous decades. The advent of statutory regulation of mortgage lenders and intermediaries will provide protection to consumers. From October 2004 the selling of mortgages will be regulated by the Financial Services Authority (FSA). Their rules will require lenders to lend responsibly and in particular lenders must “be able to show that… account was taken of the customers’ ability to repay the loan”. The FSA have a range of sanctions to enforce their rules including fines, public statements of censure, and ultimately, the removal of a firm’s permission to carry on mortgage business. The financial authorities may also have an important public information role to play in emphasising that in a low inflation environment the real value of debt payments will be eroded much more slowly than in the past. The government has recently set up a review of the mortgage market, headed by David Miles, to establish why the share of longer-term fixed rate mortgages is so low compared with many other EU countries, and whether there are obstacles to the development of a larger market for longer-term fixed-rate mortgages. A greater proportion of fixed rate mortgages might reduce the sensitivity of the housing market to movements in short term interest rates and so contain its overall instability.30 An interim report by David Miles was published in December 2003 and found little evidence that the UK mortgage market has any obvious flaws or that there are signs of anti-competitive behaviour. On the demand side the report identifies a number of factors preventing the emergence of a larger market in longer-term fixed-rate mortgages: borrowers attaching too great a weight to initial monthly payments; a poor appreciation of risk among borrowers; and the practice of mortgage lenders competing for new business to offer discounted short-term variable mortgages, against which longer-term fixed rate mortgages appear expensive. Some supply-side constraints such as capital requirements, accounting rules and legislative constraints on building societies might require a policy response to enable the market to develop properly. The final report will be published at the time of the Budget 2004. If there are indeed legislative constraints or accounting rules that discriminate against the provision of fixed-rate mortgages then policy intervention might be justified, but beyond that providing some form of tax incentives in favour of fixed-rate mortgages should be resisted as it would risk introducing other distortions. Improving housing supply The government has recently commissioned a review of issues underlying the lack of supply and responsiveness of housing in the United Kingdom, headed

© OECD 2004


55

Box 2.1. The Barker review of housing supply An interim report was published in December 2003 and identifies a range of factors which reduce housing supply including: – The key constraint on housing output is land supply, reflecting restrictions imposed by the planning system as well as the industry’s response to housing market volatility. – Local authorities have few positive incentives to encourage house building and face few sanctions if they fail to meet targets. – Consequently, competition in the industry tends to focus on land. Once land is acquired house builders have little incentive to compete for consumers or innovate. This raises the question as to whether a land tax might improve the responsiveness of housing supply. – The house building industry is thus characterised by low levels of responsiveness to demand, low levels of brownfield investment, and low levels of innovation. – Infrastructure barriers hold up construction, particularly in the South-East. – Institutional investment in property is limited. The Review suggests the government should consider tax transparent property investment trusts to encourage investment. There will be a final report with policy recommendations in spring 2004.

by Kate Barker, which will look at the role of competition, capacity and finance of the house building industry, possible fiscal instruments and the interaction of these factors with the planning system, see Box 2.1. Probably the most important policy factor that has restricted housing supply, and highlighted in the interim Barker report, is the planning system. The current planning system gives less weight to economic considerations than in most other countries, is overly complex and has too many overlapping tiers of decisionmaking. The lack of responsiveness of house building since the mid-1990s is partly due to tighter planning regulations. In particular, supply has been reduced by the direction towards brownfield and away from greenfield sites and use of Section 106 of the 1990 Town and Country Planning Act. According to the latter, local authorities can negotiate with the developer for a proportion of the planning gain to be provided to the local authority in the form of social housing, schools or other social objectives. In some areas, such as Greater London as much as half of the gain has been the target of some local authorities, with protracted negotiations adding further delays to the planning process. The government has outlined proposals for changes to the planning system (ODPM, 2002), with the aim of simplifying and speeding up the planning

© OECD 2004

56


process as well as making it more predictable. The government has also made it clear that it will be willing to intervene where local authorities do not meet housing targets in areas of high demand. There will be a new role for the Audit Commission in assessing local authorities’ performance against Regional Planning Guidance, with possible sanctions in the form of reduced funding for those authorities failing to comply. The government has also outlined proposals for four new growth areas in the South East of England to accommodate 200 000 extra homes by 2016, although these proposals are long term in nature. Assessment Reforms of the planning system which improve the responsiveness of housing supply, particularly in those areas of the country where demand is greatest are to be welcomed, not only because they would reduce the amplitude of house price cycles, but also because they are highly desirable from an overall economic performance perspective. The government should also ensure that local authorities do meet objectives for house building. Consistent with these objectives the use of “Section 106” should be reconsidered as its use clearly contradicts the aim of encouraging new development, and the lack of transparency in terms of the additional costs it imposes on developers, while the delays it introduces are also undesirable. Nevertheless, given the probable lags before there is an appreciable effect on the housing stock, other measures acting on the demand side should also be considered, not least because reducing instability of house prices may itself be an important factor in encouraging housing supply. While it is possible that variations in stamp duty might exercise a strong influence on house price developments they might have harmful effects on labour mobility. A more promising option would be the reform of the council tax to relate it more closely to current valuations of property and at the same time make it less regressive. Statutory regulation of the mortgage market by the FSA is to be welcomed. There is a need, in particular, to ensure that prudent lending requirements are not relaxed during a house price boom. In many respects the liberalised mortgage market in the United Kingdom should be seen as being of considerable benefit to the economy. EMU countries with more restrictive housing and mortgage markets would benefit from further deregulation and at the same time differences in the transmission of monetary policy across member countries would be reduced.

© OECD 2004


57

Notes 1. See Ball (2003) for further details and discussion of the UK housing system and the characteristics of the different tenures. 2. New entrants into the mortgage market are common. Many lenders offer introductory cut-price offers and, because there are often no penalties for pre-payment, borrowers often switch in order to find the best deal. 3. In many cases the cross-country variation in loan-to-value ratios reflects the ease with which the law allows lenders to repossess property in the case of default. 4. See Table 5.2 on page 51 of H.M. Treasury (2003a). 5. Mortgage equity withdrawal is borrowing that is secured on the housing stock but not invested in it, so it represents additional funds available for reinvestment or to finance consumption or investment. 6. The interest rate figures on different forms of borrowing were average rates in their categories at the end of 2002, as reported by Nickell (2003). 7. The cubic term was first used in a house price equation by Hendry (1984). Muellbauer and Murphy (2000) report that such non-linear terms explain past house price bubble episodes well, and that without such a non-linear term or dummies for spikes in the data, the equation standard error more than doubles. 8. Based on the house price index from the Office of the Deputy Prime Minister. 9. For example in the minutes of their November meeting the MPC notes: “It remained difficult to explain the continuing rise in the ratio of house prices to earnings, and hence to assess the prospects for the housing market.” 10. In May 2003 The Economist magazine predicted a house price correction of 25 per cent (Woodall, 2003). The IMF (2003) estimated that house prices exceeded their long-run equilibrium by 26 per cent in the second quarter of 2002. Conversely, Meen (2003) argues that the low level of interest rates coupled with other factors, including the inelastic supply of housing, implies that at the national level the housing market is not over-valued. 11. For assured tenancies the rent is a market rent that is freely negotiated between landlord and tenant and may be reviewed regularly. This is now the most common form of tenancy in the private sector, although in the early 1980s most tenancies were regulated. Data on assured rents are collected by the Office of the Deputy Prime Minister. 12. The Netherlands, which in many respects has a similar housing market as the United Kingdom, also experienced a combination of a rapid build up in mortgage debt, consumption-led growth and high house price inflation. However the housing market peaked in early 2000 with house prices falling in real terms recently, and consumption

© OECD 2004

58


and output growth stagnating as households have become wary about the risks of negative housing equity and high debt service since the ending of the housing boom. 13. A 1 percentage point change in short term interest rates would increase the average household’s mortgage interest payment by about 0.38 per cent of GDP, according to estimates reported on pages 26-27 of H.M. Treasury (2003a). This would be equivalent to about 0.55 per cent of disposable income. 14. Using the British Household Panel Survey, Bean (2003) estimates that roughly onethird of households had “no liquid assets to speak of”. However this estimate should be treated with caution because while the debt to income ratio implied by aggregating the survey data is similar to whole economy measures, the implied liquid asset to income ratio from the survey is less than one-third of the most obvious whole economy measures, perhaps suggesting some ambiguity in the way in which the survey question was posed. 15. Bank of England (2003), Financial Stability Review, December 2003. 16. Regressing the annual growth in real consumption on a constant and the contemporaneous change in real house prices, measured as the nominal ODPM house price series deflated by the consumers’ expenditure deflator, gives an R-squared of 0.68 compared to an R-squared of 0.54 if the contemporaneous change in real personal disposable income is used. 17. The estimates reported in Annex 2.A1 suggest that a 1 per cent change in housing wealth raises consumption by between 0.12 and 0.3 per cent in the short-run. In all cases these short-run effects are substantially larger than the estimated long-run effects. 18. The Interlink model does not include housing or financial wealth, so instead the effect of the shock to house prices on consumption has been calculated according to the estimated consumption equation reported in Annex 2.A1 and this shock has been applied to the Interlink consumption equation. 19. While asset price adjustments are rarely smooth, for forecasting purposes this remains a convenient assumption given the impossibility of reliably predicting the timing of any future abrupt correction. This also serves to emphasise the uncertainty of both the magnitude and especially the timing of the correction considered in the simulation exercise. 20. Although the profile is very different, the level of house prices is approximately the same in the two scenarios by 2007. 21. Past or present members of the MPC who have elucidated the argument against a more pro-active monetary policy against asset prices include Vickers (1999), Allsopp (2002), King (2002), Nickell (2003), Bell (2002) and Bean (2003). 22. Stamp duty in the United Kingdom is currently paid at 4 rates (0, 1, 3 and 4 per cent) depending on the price of the property. 23. Residential mobility rates are here defined as the number of yearly owner moves divided by the number of owners. 24. Properties are banded according to an assessment of their market value (as at April 1991), with local authorities determining the rate levels levied on these bands subject to a constraint regarding the ratio of tax paid in each band relative to a reference band. Thus, for example, within any particular local authority the absolute level of council tax on a property valued at £300 000 will only be 2½ times that levied on a

© OECD 2004


25.

26.

27. 28.

29.

30.

59

property valued at £30 000 (rather than 10 times as would be implied by a constant tax rate). See Adam and Kaplan (2002) for further details. Reforming the council tax in this way would, however, reduce the discretionary powers of local authorities to raise revenue, suggesting that it would need to take place in the context of a review of local government finances. On the other hand, reforming the council tax could have other significant benefits, such as easing unemployment and poverty traps (Muellbauer and Cameron, 2000). If the council tax were reformed in a revenue-neutral way as a constant tax rate on the value of all household property with frequent revaluations, and some approximation based on a national index of prices applied in intervening years, then a 20 per cent increase in house prices would automatically reduce real personal disposable income by nearly ½ a per cent. A higher tax rate of 3 per cent is paid on properties over a threshold valuation of DKK 3 040 000 (about EUR 400 000). When the current Danish government took office in late 2001, it introduced a nominal ceiling on the property value tax, so that an increase in the property value could not lead to higher tax payments for the individual homeowner. The recent OECD Survey of Denmark argued that a better economic outcome could be obtained by adjusting property tax payments in line with inflation while gradually reducing the high marginal taxes on earned income instead. Simulations of the European Commission’s QUEST model suggest that across all European countries for a shock to consumption the automatic stabilisers operate most strongly in the case of Denmark and Sweden (EC, 2001). The European Mortgage Finance Agency has recently announced plans to create a new agency that would aim to establish a European-based version of the US-based Fannie Mae or Freddie Mac (Thomas, 2003). It would act as a source of funding for mortgage lenders while leaving the business of selling loans to lenders in each national market. If it were to be given some form of guarantee by the European Union, then it is claimed that the option of a long-term fixed rate mortgage without redemption penalties could be provided to European consumers. However, it remains to be seen whether the EU will provide such guarantees, and few institutions based in the United Kingdom have so far shown an interest in the new venture.

© OECD 2004

60


Bibliography Adam, S. and G. Kaplan (2002), “A Survey of the UK Tax System”, Institute for Fiscal Studies, Briefing Note, No.9, available from www.ifs.org.uk/taxsystem/taxsurvey.pdf. Allsopp, C. (2002), speech delivered to the Conference on ’Policy Rules – The Next Steps’, Cambridge University, Cambridge on 19 September, available from www.bankofengland. co.uk/Links/setframe.htm. Ball, M. (2003), RICS European Housing Review 2003, RICS Policy Unit and RICS Residential Property Faculty, www.rics.org/resources/research/. Bank of Denmark (Danmarks Nationalbank) (2003), MONA – en kvartalsmodel af dansk økonomi, documentation for the quarterly macroeconomic model of the Danish Central Bank. Bank of England (2000), “Economic Models at the Bank of England – September 2000 Update”, pp. 29-30, available from www.bankofengland.co.u. Bank of England (2003), Financial Stability Review, December, 2003. Bean, C. (2003), “Assets Prices, Financial Imbalances and Monetary Policy: Are Inflation Targets Enough?”, July, available from www.bankofengland.co.uk/Links/setframe.html. Bell, M. (2002), “Six monfhs on the MPC: A reflection on Monetary Policy“, speech to the CBI South East in Crawley, Sussex on Monday 9 December, available from www.bankofengland.co.uk/Links/setframe.htm. Cox, P., J. Whitley and P. Brierly (2002), “Financial Pressures in the UK Household Sector: Evidence from the British Household Panel Survey”, Bank of England Quarterly Bulletin, Winter, available from www.bankofengland.co.uk/qb/qb020405.pdf. EC (European Commission) (2001), “Public Finances in EMU – 2001”, European Economy, No. 3. FSA (Financial Services Authority) (2003), Financial Risk Outlook 2003, www.fsa.gov.uk/pubs. Girouard, N., and S. Blondal (2001), “House Prices and Economic Activity”, OECD Economics Working Paper, No. 279, www.oecd.org/. Hamilton, R. (2003), “Trends in Household Secured Debt”, Bank of England Quarterly Bulletin, Autumn. Hendry, D.F. (1984), “Economic Modelling of House Prices in the United Kingdom”, in Econometrics and Quantitative Economics, D.F. Hendry and K.F. Wallis (eds.), Basil Blackwell, Oxford. H.M. Treasury (2003a), “Housing, Consumption and EMU”, EMU study to inform the assessment of the five economic tests, www.hm–treasury.gov.uk/documents/the_euro/euro_index_index.cfm. H.M. Treasury (2003b), “Fiscal Stabilisation and EMU”, EMU study to inform the assessment of the five economic tests, www.hm-treasury.gov.uk/documents/the_euro/euro_index_index.cfm. IMF (International Monetary Fund) (2003), United Kingdom, Selected Issues, February.

© OECD 2004


61

King, M. (2002), speech to the London School of Economics on Tuesday 19 November 2002, available from www.bankofengland.co.uk/Links/setframe.html. Malpezzi, S. and D. Maclennan (2001), “The Price Elasticity of Supply of New Residential Construction in the United States and the United Kingdom”, Journal of Housing Economics, No. 10. Meen, G. (1996), “Ten Propositions in UK Housing Macroeconomics: An Overview of the 1980s and Early 1990s”, Urban Studies, No. 33 (3). Meen, G. (2003), “Can House Prices Remain so High?,”Economic Outlook, Vol. 27, No. 2, January, Oxford Economic Forecasting and London Business School. Muellbauer, J., and G. Cameron (2000), “Five Key Council Tax Reforms and Twelve Reasons to Enact Them”, New Economy, June, Vol. 7, No. 2. Muellbauer, J., and A. Murphy (2000), “Booms and Busts in the UK Housing Market”, March, available from www.housingoutlook.co.uk. Muellbauer, J. (2003a), “The UK and the Euro: The Role of Asymmetries in Housing and Credit Markets”, Submission to the EMU study to inform the assessment of the five economic tests. Muellbauer, J. (2003b), “Housing, Credit and the Euro: the Policy Response”, Economic Outlook, Vol. 27, No. 4, July, Oxford Economic Forecasting and London Business School. Nickell, S. (2003), “House Prices, Household Debt and Monetary Policy” , Quarterly Bulletin, Vol. 43, No. 1, Spring, Bank of England. Van den Noord, P. (2003), “Tax Incentives and House Price Volatility in the Euro Area; Theory and Evidence”, OECD Economics Department Working Paper, No. 356, www.oecd.org/ OECD (2003), Economic Survey of Ireland, Vol. 2003/9, July, OECD, Paris. ODPM (Office of the Deputy Prime Minister) (2002), “Sustainable Communities: Delivering through Planning”, July, www.odpm.gov.uk/ van Ommeren, J. and M. van Leuvensteijn, (2003), “New Evidence of the Effect of Transaction Costs on Residential Mobility”, CPB Discussion Paper, No.18, May, http://ideas.repec.org/ p/cpb/discus/18.html. Swank, J., J. Kakes and A. Tieman, (2002), “The Housing Ladder, Taxation and Borrowing Constraints”, DNB Staff Reports, No.9 available from: http://netec.mcc.ac.uk/WoPEc/data/Papers/ dnbwormem688.html. Thomas, R. (2003), “Bringing Europe’s Mortgage Markets under one Roof”, Financial Times, November 21. Vickers, J. (1999), “Monetary Policy and Asset Prices“, speech delivered to the Money, Macro and Finance Group 31st Annual Conference, Oxford University, 22 September, available from: www.bankofengland.co.uk/Links/setframe.html. Woodall, P. (2003), “Close to Bursting. A Survey of Property”, The Economist, 31 May, The Economist Newspaper Limited, London.

© OECD 2004


62

Annex 2.A1.

Consumption equation including housing wealth

This annex reports an estimated equation explaining real consumption (C) in terms of real disposable income (Y), real housing wealth (HW), real financial wealth (FW) and nominal short – term interest rates (IRS). The equation is estimated on quarterly data over a sample period from 1981 Q1 to 2003 Q1. The sample was not extended back into the 1970s both because of problems of ensuring a consistent data set and to avoid instability in the equation responses due to financial market deregulation.

∆InC =

–0.0114 + 0.1983 ∆Y – 0.0874 ∆IRS–1 – 0.0791 ∆IRS–2 – 0.0765 ∆IRS–3 (–2.0)

(4.6)

(–1.5)

(–1.4)

(–1.3)

+0.1343 ∆InHW + 0.1555 ∆InHW–2 + 0.0188 ∆InFW–1 (4.2)

(4.2)

(2.3)

–0.1494 In(C/Y)–1 + 0.0039 In((HW + FW)/Y)–1 (–4.7)

(1.3)

Rsqd-adjusted = 0.55, t-statistics reported in brackets, Standard error of regression = 0.0048 LM – test for up to second order serial correlation = 1.39 (p-value =0.25), Ramsey RESET test = 1.93 (p-value = 0.15), Jarque-Bera test for normality of residuals = 1.78 (p-value =0.41), White heteroskedasticity test = 1.22 (p-value =0.27). The restriction that financial and housing wealth have the same long-run effect on consumption is readily accepted at the 5 per cent significance level and is imposed in this equation, although the freely estimated long – run effect of housing wealth is higher than for financial wealth. Of particular interest, as discussed in the main text, are the relative magnitudes of the short and long-run responses to housing wealth which are compared to the responses from similar equations reported in recent empirical work by both the Treasury and Bank of England (Table 2.A1.1). In all cases the short-run response is much larger than the eventual long-run response. Some caution must, however, be attached to the long-run estimates of housing wealth from the current equation above, which is not well determined.

© OECD 2004


63

Table 2.A1.1. Response of consumption to housing wealth1 Consumption equation

Short-run response Per cent

Equation reported in this annex Bank of England H.M. Treasury

2.5 (after 3 quarters) 1.2 (after 1 quarter) 3.0 (“in short-run”)

Long-run response Per cent

0.3 0.5 1.0

1. Single equation effect following a 10 per cent increase in housing wealth. Source: H.M. Treasury (2003a), “Housing, Consumption and EMU”, pp 71-73, EMU study, available from www.hm-treasury.gov.uk, and Bank of England (2000), “Economic Models at the Bank of England – September 2000 Update”, pp. 29-30, available from www.bankofengland.co.uk.

© OECD 2004

III. The fiscal challenge: complying with the fiscal rules while raising standards in health and education Introduction The United Kingdom has endeavoured to increase the quality of public services in three strategic areas where it lags behind countries with a similar degree of economic development: public infrastructure, education and health. Raising quality of services involves more public spending. This ambitious spending plan will have to be executed without compromising a fiscal framework that has served the United Kingdom well so far and is mainly based on a “golden rule” that seeks to prevent a decline in public net wealth in the long-run while providing flexibility over the cycle. This framework allows the financing of new public infrastructure subject to a “sustainable investment rule” which sets a target for net public sector debt that is comfortably met at present. However, the increase in current expenditure will have to come either from increased taxation or slower growth of expenditure in the long run. The government has already increased taxes and has set very strict targets for spending outside health and education. The government also started from a strong budgetary position that could accommodate part of the spending increases on health and education. However, the surpluses achieved in 1999-2000 seem to have had an exceptional character, with buoyant tax receipts linked to the equity price bubble. Recent budgetary developments indeed suggest that the underlying position of public finances is weaker than originally expected in Budget 2003. Going forward, an important strategic variable is the speed of the build up in priority spending. It could be reduced on several grounds: firstly, cost-effectiveness needs to be ensured, to lock in improved performance, rather than higher costs; secondly, more time to pilot and implement promising innovations in health care would be available; and finally, potential strains on a fiscal framework that has been very successful so far would be diminished. This chapter will first assess how much the government deficit has departed from the original plans and evaluate the consequences for the government’s room for manoeuvre. It then considers how best to meet the challenge of adhering to the government’s fiscal rules and code while at the same time raising

© OECD 2004

66


standards in health and education. An in-depth examination suggests that the existing room for budgetary manoeuvre is less than expected and that securing increases in the quality of services may call for a somewhat less rapid growth in spending in the future. Moreover, there seems to be scope for introducing user charges which do not compromise equity concerns and introducing activity-based funding for hospitals and incentive pay for doctors would probably also boost productivity. The aggregate fiscal position An update of official fiscal projections in the (December) 2003 Pre-Budget Report (PBR) suggests that the government deficit, on a Maastricht basis, is expected to be around 3¼ per cent of GDP in fiscal year 2003, compared to the (April) 2003 Budget forecast of 2½ per cent of GDP (H.M. Treasury, 2003a, 2003b). This overrun does not a priori stem from cyclical weakness as the macroeconomic projections underlying the Budget forecast of public finances are likely to be realised. While expenditure overruns appear to be at least as important as revenue weakness in the fiscal overshoot for 2003, tax revenues have been unexpectedly weak compared to the Budget projections made in 2001 and 2002. For example, general government current receipts in fiscal year 2002 were 2½ per cent of GDP lower than expected at the time of the 2001 Budget (Table 3.1). One indication of the importance of exceptional factors is that the Treasury’s estimate of the cyclically-adjusted current balance in fiscal year 2002 is 1½ per cent of GDP lower than projected at the time of the 2001 Budget, despite current expenditure being in line with the projection and the absence of any major change in tax rates implemented during fiscal year 2002. In order to fund a sustained increase in health expenditure and to maintain sound public finances, the 2002 Budget announced tax increases. These were mainly in the form of a rise in employees’ and employers’ national insurance contributions, to be mostly implemented in fiscal year 2003, bolstering tax revenues by about 0.6 per cent of GDP (Box 3.1). An important reason for this revenue weakness is that tax revenues in the United Kingdom, particularly those from direct taxes on companies and taxes on financial and capital transactions, appear to be more sensitive to movements in equity prices than in most other OECD countries.1 Using independent estimates as well as scaling-up estimates reported in the 2002 Pre-Budget Report suggests that the 40 per cent fall in equity prices2 between fiscal year 2000 and 2002 might have reduced tax revenues from capital taxes, stamp duty and corporation tax by between ¾ and 1 percentage point of GDP, although it is likely that some of this effect only comes through with a lag and so is damping receipts in 2003 also.3 Equity markets have a direct effect on corporation tax revenues through their effect on capital gains, although the strength of any broader effects on financial sector profitability relative to other factors is uncertain. In addition the buoyant

© OECD 2004

The fiscal challenge: complying with the fiscal rules...

67

Table 3.1. Public sector finances: selected summary indicators and official projections As a per cent of GDP1 2001/2

2002/3

2003/4

2004/5

2005/6

2006/7

2007/8

Flows Current receipts Budget 2001 Budget 2002 Budget 2003 Pre-Budget Report 2003

40.2 39.1 38.8 38.8

40.1 38.7 37.6 37.6

39.8 39.9 38.6 38.1

39.9 40.2 39.2 38.9

39.8 40.4 39.8 39.5

.. 40.5 40.1 39.9

.. .. 40.4 40.2

Current expenditure Budget 2001 Budget 2002 Budget 2003 Pre-Budget Report 2003

37.1 36.7 36.5 36.5

37.3 37.1 37.4 37.3

37.5 37.9 38.1 38.5

37.5 38.2 38.1 38.2

37.5 38.5 38.4 38.6

.. 38.5 38.4 38.5

.. .. 38.5 38.6

Depreciation

1.3

1.4

1.3

1.3

1.3

1.3

1.3

Gross investment

2.7

2.8

3.3

3.6

3.7

3.7

3.8

Net investment2

0.9

1.0

1.6

1.9

2.0

2.1

2.2

Surplus on current budget Budget 2003 Pre-Budget Report 2003

1.0 0.9

–1.1 –1.1

–0.8 –1.7

–0.1 –0.7

0.2 –0.4

0.4 0.0

0.6 0.3

Average surplus since 1999-2000 Budget 2003 Pre-Budget Report 2003

1.8 1.7

1.1 1.0

0.7 0.4

0.6 0.3

0.5 0.2

0.5 0.1

0.5 0.2

Net lending3 Budget 2003 Pre-Budget Report 2003

0.0 0.0

–2.3 –2.1

–2.4 –3.3

–2.1 –2.6

–1.9 –2.4

–1.7 –2.1

–1.7 –2.0

Cyclically-adjusted balances Surplus on current budget Budget 2001 Budget 2002 Budget 2003 Pre-Budget Report 2003

1.4 1.0 0.9 0.6

1.1 0.5 –0.5 –0.6

0.6 0.6 0.2 –0.8

0.7 0.7 0.5 –0.1

0.7 0.6 0.4 –0.1

.. 0.7 0.4 0.0

.. .. 0.6 0.3

Net lending Budget 2001 Budget 2002 Budget 2003 Pre-Budget Report 2003

0.3 –0.3 –0.1 –0.3

–0.3 –0.8 –1.7 –1.5

–1.1 –1.1 –1.5 –2.4

–1.1 –1.2 –1.4 –2.0

–1.1 –1.4 –1.7 –2.2

.. –1.5 –1.7 –2.1

.. .. –1.7 –2.0

End period stocks Public sector net debt General government gross debt

30.2 37.9

30.9 37.9

32.8 39.3

33.8 40.2

34.6 40.8

35.1 41.1

35.4 41.4

0.0

–1.1

–1.4

–0.7

–0.2

0.0

0.0

Memorandum item: Output gap

1. Pre-Budget Report, December 2003, unless otherwise noted. 2. Equals gross investment minus net asset sales less depreciation. 3. Indicator relevant to the Stability and Growth Pact. Source: H.M. Treasury.

© OECD 2004


68

Box 3.1.

The 2002 and 2003 Budgets and the 2003 Pre-Budget Report

In terms of fiscal impact, the main measures announced in the 2002 Budget only took effect at the time of the 2003 Budget with the net effect of raising revenues (relative to an indexed base) by £6.1 billion in fiscal year 2003, equivalent to about 0.6 per cent of GDP. The main measures included: – An increase in national insurance contributions from employees, employers and the self-employed by £3.5, £3.9 and £0.45 billion, respectively, in fiscal year 2003, together equivalent to about 1.0 per cent of the total wage bill for the whole economy. – The introduction of the child tax credit and working tax credit, at a cost of £0.5 billion in fiscal year 2002 and £2.5 billion in fiscal year 2003. This measure will raise the incomes of those in work on the lowest incomes, as discussed in Chapter IV. – Various tax measures aimed at reducing tax-avoidance and changes in North sea oil taxation which increased revenues by £1.4 billion and £0.45 billion, respectively, in fiscal year 2003. In terms of their macroeconomic impact the measures announced in the 2003 Budget were relatively minor, with the net impact of the changes (relative to an indexed base) costing about £500 million (about half of 0.1 percentage point of GDP). This was roughly offset by the fiscal impact of other measures, mainly involving higher tax revenues from anti-avoidance measures, announced between the 2002 and 2003 Budgets. Among the measures in the 2003 Budget having the greatest fiscal impact were the postponement of the rise in fuel duties (given the uncertainties surrounding the oil market due to the Iraq conflict) and further measures designed to reduce tax avoidance. The contingency provision was also increased to £3 billion to ensure that resources were available to cover the full cost of military obligations in Iraq and £240 million was allocated towards the costs of humanitarian assistance. The main fiscal initiative announced in the 2003 Pre-Budget Report was an increase in Child Tax Credit, which will cost about £0.9 billion annually (just under 0.1 per cent of GDP) and will be implemented in April 2004, mainly benefiting low-income families with children. To pay for continuing international commitments in Iraq, £2 billion of out of the special reserve initially allocated in Budget 2003 will be carried forward into fiscal year 2003, and a further £500 million will be added to take the total reserve to £2.5 billion. A further £300 million will be added in fiscal year 2004 as a prudent allowance against continuing commitments.

stock market of the late 1990s allowed companies to reduce their (tax-deductible) contributions to pension schemes and so boosted tax revenues, an effect which may now be reversed as companies are obliged to increase contributions in order to reduce pension funding gaps. A further indirect effect from the buoyant equity market arose from the effect that strong growth of financial companies profits had on wage bonuses going to higher rate taxpayers. This largely explained the addi-

© OECD 2004


69

tional £3¼ billion (0.3 per cent of GDP) of unexpected income tax revenues in fiscal year 2000 (H.M. Treasury, 2002a). Summing these effects suggests that, in the absence of a sustained improvement in the stock market (relative to money GDP), tax revenues would be reduced by at least 1 per cent of GDP relative to the position in 2000. Will the fiscal rules be met? The 2002 Spending Review set public expenditure plans for the three years to fiscal year 2005, and, exceptionally, five years in the case of health expenditure to fiscal year 2007. Total current expenditure is set to rise by around 12 per cent in real terms (evaluated relative to the expected change in the GDP deflator) between fiscal years 2002 and 2005, with public sector net investment rising from 1 per cent of GDP in fiscal year 2002 to 2¼ per cent of GDP in 2007. Much of the extra spending is in the priority areas of health and education, with National Health Service (NHS) spending planned to rise on average by 7 per cent per annum in real terms over the five years to fiscal year 2007 and education expenditure rising by 6 per cent per annum in real terms to fiscal year 2005. Official fiscal projections The fiscal projections in the 2003 Pre-Budget Report suggest that, despite these increases in expenditure, the “golden rule” (Box 3.2) will continue to be met. However, the margin by which it is projected to hold has been reduced since the 2003 Budget with the average current budget balance since the beginning of the cycle (assumed to be in fiscal year 1999) projected to remain in surplus by only 0.2 per cent of GDP in fiscal year 2005 when the output gap is projected to be closed (Table 3.1). Looking beyond the end of the current cycle (when output is assumed to follow trend), the PBR projections suggest that the golden rule will also be met, with the current surplus projected to rise from 0.0 to 0.6 per cent of GDP between fiscal years 2006 and 2008. With net government investment also projected to increase by nearly ½ per cent of GDP between fiscal years 2003 and 2005 to 2 per cent of GDP, net borrowing is expected to fall only gradually below 2½ per cent of GDP by fiscal year 2005. The deficit ratio would only exceed the 3 per cent Maastricht Treaty limit in fiscal year 2003, corresponding to 2.4 per cent in cyclically adjusted terms, but thereafter the deficit falls below the Maastricht Treaty limit and would probably also remain below those of most other major European economies. Moreover, general government debt, at less than 40 per cent of GDP, would stay well below that in other major European economies and well below the 60 per cent limit set in the Maastricht Treaty. The improvement in the fiscal position in the PBR projections occurs mainly because of a rise in the aggregate tax to GDP ratio of 2.1 percentage points

© OECD 2004


70

Box 3.2.

The fiscal framework

The “Code for Fiscal Stability” guides the formulation and implementation of fiscal policy based on the five principles of transparency, stability, responsibility, fairness and efficiency (H.M. Treasury, 1998). Consistent with this code the government has repeatedly emphasised the importance of adhering to two fiscal “rules” in the conduct of fiscal policy: – The golden rule: over the economic cycle the government will borrow only to invest and not to fund current spending, implying that on average over the cycle the current balance will be in surplus. – The sustainable investment rule: over the economic cycle, the ratio of public sector net debt to GDP will be held at a stable and prudent level, defined to be below 40 per cent of GDP. The definition and dating of the economic cycle are important practical concerns for the implementation of these rules, although this inevitably involves a degree of judgement. The Treasury uses a variety of evidence, not just national accounts data but including also survey evidence, to determine “on-trend” points (H.M. Treasury, 1999). On this basis the current cycle is dated as having begun in fiscal year 1999, with output initially exceeding trend, before falling below it in 2001. The current cycle will then have ended when GDP rises to reach trend again, which according to the Budget 2003 projections will be towards the end of fiscal year 2005. Also the OECD projections suggest that the output gap will be closed by the end of 2005. There are differences in definitions and focus as compared to the EU Treaty and Stability and Growth Pact commitments. In particular: – the UK fiscal rules apply to the whole public sector, whereas the Treaty deficit and debt criteria only apply to general government; – the current balance excludes capital spending, while the Treaty focuses on the budget deficit; – the UK debt measure is net of liquid assets, whereas the Treaty measure uses gross debt; – the UK fiscal rules apply over the course of a full economic cycle rather than on a year-to-year basis as is the case for the 3 per cent of GDP Maastricht deficit limit. Cleary the two sets of fiscal rules are based on different concepts. In particular, the UK deficit could exceed the Maastricht Treaty’s 3 per cent of GDP limit in any given year, and yet be consistent with the Golden Rule and the Sustainable Investment rule. A potential weakness of the golden rule is that judging it according to the cumulative balance incurred since the beginning of the cycle may be inappropriate if revenues or spending earlier in the cycle are in some way “exceptional”. This may be particularly relevant in current circumstances where revenues in 1999 and 2000 were boosted by the exceptional strength of the equity market. However, in compliance with the spirit of the Code for Fiscal Stability, fiscal policy is set on a forward-looking basis to be consistent with the Golden Rule in the medium term, and beyond the end of current cycle.

© OECD 2004


71

Figure 3.1. Tax-to-GDP ratio As a per cent of GDP 40

40 Pre-Budget Report 2003 Projections

39

39

38

38

37

37

36

36

35

35

34

34

33

33

32

1980

1985

1990

1995

2000

2005

32

Source: H.M. Treasury, Budget 2003.

between fiscal years 2003 and 2007 (Figure 3.1), and to a lesser extent the easing of one-off spending pressures (accounting for about 0.2 per cent of GDP) due to the UK international commitments affecting fiscal years 2002 and 2003. Almost half of the increase in the tax to GDP ratio is attributable to non-North sea corporation tax which is expected to rise from the historically low level of 2.3 per cent of GDP in fiscal year 2003 to 3.2 per cent by fiscal year 2007 (Figure 3.2), still around 0.4 percentage point lower than in fiscal year 1999 at the peak of the stock market boom. A small part of this increase can be accounted for by the recovery of equity markets through 2003, which are estimated to raise receipts from capital gains tax, corporate tax paid by life assurance companies and stamp duty on shares by about 0.1 per cent of GDP from fiscal year 2004. However, despite the assumption that in future equity prices only rise in line with nominal GDP, the profitability of financial companies, and hence the corporate tax take in relation to GDP, is projected to rise as a consequence of other factors affecting profitability such as a rise in merger and acquisition activities and interest rates spreads. The income tax to GDP ratio is also forecast to rise by 0.8 percentage point between fiscal years 2003 and 2007, mainly as a consequence of real fiscal drag (adding about 0.5 percentage point) and an allowance for the expected improvement in financial company profits on bonus payments. An expansion of tax compliance measures is also expected to boost the aggregate tax to GDP ratio by 0.2 percentage point by fiscal year 2006.

© OECD 2004


72

Figure 3.2. Non-North Sea corporation tax receipts1 % of GDP

Index 1980 = 100

300

4.5 Projections Pre-Budget Report 2003

4.0

250

3.5 3.0

200 2.5 2.0

150

1.5 1.0

Non-North Sea corporation tax receipts (left scale) FTSE all share index normalised on GDP (right scale)

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006

100

1. Financial years. Source: OECD, Inland revenue, Office for National Statistics and H.M. Treasury, Pre-Budget Report 2003.

OECD fiscal projections Both the OECD and 2003 PBR projections suggest that the government’s golden rule will be met over the current cycle, although in both cases the safety margin is not large: the OECD projections suggest that the average current balance surplus since the start of the cycle may only amount to 0.1 percentage point by 2005.4 An important difference, however, between the two projections is that the OECD’s projections suggest no improvement in either the current budget balance or the overall deficit on a Maastricht basis. The main reasons for this is less of a pick-up in the aggregate tax ratio as the output gap, which is estimated to be initially smaller than the 2003 PBR estimate, closes: on the OECD projections the tax to GDP ratio rises by 0.9 percentage point between 2002 and 2005, whereas according to the PBR projections it rises by 1.7 percentage points. The current budget balance remains in substantial deficit, at about 1¾ per cent of GDP, until the cycle finishes in 2005, and the government deficit, on a Maastricht basis, would remain close to 3 per cent (Figure 3.3). The OECD’s econometric estimates, presented in Annex 3.A1, confirm the sensitivity of the corporate tax take to both the state of the cycle and the stock market, and suggest that a closing of the current output gap will raise corporate tax revenues by 0.4 percentage point of GDP between 2002 and 2005, slightly less than the increase projected in the 2003 PreBudget Report. Revenues from income tax, which the OECD’s forecast projects to

© OECD 2004


73

Figure 3.3. OECD projections of general government finances As a per cent of GDP 5

5 Projections

4

4

3

3

2

2

1

1

0

0

-1

-1

-2 -3 -4

-2

Actual surplus on current budget Cyclically-adjusted current balance Net lending (Maastricht basis) Average current surplus since start of cycle

1999

2000

2001

-3 2002

2003

2004

2005

-4

Source: OECD secretariat.

remain constant as a percentage of GDP between 2002 and 2005 account for a bigger difference in the buoyancy of official and OECD projections. There is also a smaller contribution to this difference due to projected revenues from “other taxes” (including council tax). The absence of any improvement in the current balance, according to the OECD projections, suggests that fiscal actions, either to slow spending or raise revenues, are required to keep the fiscal projections on track to meet the golden rule over the next cycle. Moreover, it would be advisable to start implementing these changes during the current upswing rather than postpone measures and thereby risk reinforcing the downswing of the next cycle. In evaluating the risks to the medium-term fiscal projections, the underlying potential growth assumptions are particularly important. The Treasury revised up the central assumption of potential output growth between the 2002 and 2003 Budgets from 2½ to 2¾ per cent per annum,5 although as customary it uses a lower “deliberately cautious” assumption, in this case 2½ per cent per annum, for projecting the public finances.6 Moreover, on the basis of recent historical estimates of trend productivity growth, following revisions to GDP data, the Treasury have greater confidence in these estimates. However, the Treasury’s cautious assumption of 2½ per cent potential growth to fiscal year 2006 corresponds to what the OECD Secretariat regards as a central case (as discussed in greater detail in Annex 3.A2). Viewed in this light the assumptions underlying the Budget pro-

© OECD 2004


74

jections appear realistic, but might provide a smaller cushion than assumed by the Treasury. Rising public spending on health care and education Improving public services has been a key priority of the current government since it took office in 1997, and to this end it was seen as necessary to rapidly raise public spending on health care and education, which has been low by international comparison (Figure 3.4). In health care, the number of practicing doctors per head of population is relatively low, survival rates after cancer operations have been among the lowest in Western Europe,7 for many years patients have had to wait much longer for medical treatment than in other OECD countries, and public dissatisfaction with the health service has been widespread.8 In education, notably the number of young people getting vocational skills has been low compared with many other countries (Chapter IV). Raising public spending on these services appears to be necessary, but providing more public funding is only one element determining the quality of government services. Improving the organisation and incentives for delivering services and allowing individuals to spend more privately, are also important. Assessing the effectiveness of such an ambitious increase in public health and education spending is fraught with difficulties because of time lags and the lack of good indicators of performance. There are nonetheless signs that the Figure 3.4. Health and education expenditure in international perspective As a per cent of GDP B. Education expenditure, 2000 1

A. Health expenditure, 2001 IRE

NLD

FIN

Public

UK

Private

JPN

ITA

Public

JPN

Private

IRE

ITA

UK

SWE

DEU

NLD

FIN

AUS

FRA

FRA

CHE

CAN

CAN

DEU

AUS

CHE

SWE

USA

USA

0

2

4

6

8

10

12

14

16

0

2

4

6

8

10

1. Total of all levels of education from primary school to university. Public includes government expenditure directly for education institutions and subsidies for private expenditure for education institutions. Source: OECD Health Data and OECD Education Database.

© OECD 2004


75

speed with which the increase in spending is being implemented has perhaps hit a limit. The gains from successive increases in expenditure may be declining because of wage pressures and limits on the ability of organisations to absorb rapid spending increases efficiently. National accounts data suggest that the growth in spending on public consumption is resulting in price increases well above inflation, whereas the volume increases are limited. As discussed later however, the volume measures used by the Office of National Statistics tend to underestimate growth in public service provision and quality, and consequently the public consumption deflator in the national accounts tends to overestimate inflationary pressures. It is therefore commendable that the National Statistician has asked Sir Tony Atkinson to undertake a review of the future development of measures of government output, productivity and associated price indices so as to advance methodologies.9 Moreover, there is nothing wrong a priori with the deflator for public consumption growing faster than the GDP deflator in the long run, as public consumption is concentrated on services with relatively slow measured productivity growth. Nevertheless, the difference between the public consumption deflator and the GDP deflator has been unusually high during recent years (Figure 3.5) coinciding with the rise in spending taking place since 1998. The strength of inflationary pressures is also highlighted by the fact that in late 2003, the year-on-year increase in public sector average earnings was nearly double that in the private sector. Figure 3.5.

Speed limits have been reached for public spending Year-on-year percentage change

14 12 10

8 A. General government final consumption expenditure Price Volume

B. Public and private sector earnings 7

Public Private HICP

6

8

5 GDP deflator

6

4 4 3

2

2

0

1

-2 -4

1994

1996

1998

2000

Source: Office for National Statistics.

© OECD 2004

2002

1994

1996

1998

2000

2002

0

76


A first condition for increased spending to improve services is that more spending succeeds in buying more input rather than pushing up wages and costs for scarce resources. A second condition for service improvements to materialize is that the organisations providing services are able to turn more input into more output and better outcomes. Buying more inputs… Examining the labour markets for health and education professionals, there are signs of capacity constraints for doctors, but less so for nurses and teachers. Since 1998, medical practitioners have had cumulative increase in average real earnings per hour that are 13 percentage points greater than the economywide average (Figure 3.6), while earnings growth for nurses and teachers has been more aligned with that of other occupations. Three-year agreements concluded with all NHS staff during 2003 will mean that their earnings will grow ahead of projected earnings growth in the private sector. For doctors working in hospitals, average career earnings will increase by 25-35 per cent over the coming three years. The rest of the NHS staff will have a general 10 per cent pay increase over the next three years and many will be upgraded to higher-paying job functions in return for changes in organisation and work practices intended to enhance productivity and enable delivery of better services for patients. For teachers, increasing the average working week by over three hours during the last decade (Figure 3.6) has helped to expand labour supply in a context where the average working week for all professions in the economy has declined by half an hour, but for medical practitioners the average working week has contracted over recent years. As with earnings increases, recruitment shortages are pronounced for doctors.Staff numbers have increased by over 15 per cent for medical doctors and dentists, for qualified nurses and midwives, and for other therapeutic, support and infrastructure staff since 1997 (Table 3.2). Following this growth, it has been difficult to recruit doctors as hospital consultants, and the planned increase of 7 000 persons from 1999 to March 2004 was only halfway achieved by September 2002. For general medical practitioners, 1 535 of the planned increase of 2 000 had been achieved by June 2003. For nurses, the recruitment plans originally set for March 2004 and 2005 have already been met.10 An important reason why it has been possible to raise staff so fast is that the UK public sector is in a favourable position to attract foreign professionals because of the absence of language barriers and the historical links with developing countries. More than a third of the growth in medical staff has come from persons qualified outside the European Union, so that now one in four medical doctors in UK public health care come from countries like South Africa and India (Table 3.2). Another important reason is that a range of non-wage issues have been addressed in order to make jobs more attractive, such as improving childcare availability and tackling violence

© OECD 2004


77

Figure 3.6. Earnings and working time in health care and education Percentage change from previous year

Percentage change from previous year

A. Average real hourly earnings including overtime 10

All-occupations

Nurses

Medical-practioners

Teachers

10

5

5

0

0

-5

-5 1992

44

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

B. Average weekly hours worked including overtime

34

42

32

40

30

38

28 All-occupations (left scale) Medical-practioners (left scale)

36 1992

1993

1994

1995

1996

1997

Nurses (left scale) Teachers (right scale)

1998

1999

2000

2001

26 2002

2003

Source: Office for National Statistics, New Earnings Survey.

against health care staff (Department of Health, 2003a). Moreover, recruitment difficulties are not equally severe in different parts of the country and, despite the current “London supplement” in public sector pay, vacancy rates are typically higher in the capital (H.M. Treasury, 2002b). In general, the government’s strategy for the public sector is to target these regional and local recruitment difficulties with greater differentiation in pay.11 While these policies have successfully facilitated rapid recruitment of for example nurses, the difficulties of recruiting doctors illustrates the speed limits for health care expansion: By fiscal year 2007, the United Kingdom is expected to spend 9½ per cent of its GDP on health care (up from below 7 per cent in fiscal year 1998) and thereby reach the current level in countries like Netherlands and France, but even then there will still be only about

© OECD 2004


78

Table 3.2. Employment growth in the public sector Total growth Per cent 1992-97

Private sector Public sector

9.8

Level in thousands

1997-02

3.5

2002

23 610

–14.3

6.9

5 298

NHS England

–0.4

15.6

1 224

Medical doctors and dentists Growth decomposed by country of qualification:1 United Kingdom Elsewhere in the European Union Outside the European Union

14.1

15.3

103

.. .. ..

8.5 0.4 6.4

73 5 25

Qualified nurses, midwives, etc.

–2.2

15.3

368

Other therapeutic and infrastructure staff

–1.2

15.8

753

Note: Numbers for the European Union also include Iceland, Norway and Liechtenstein that are part of the European Economic Area as well as Switzerland. 1. Based on the assumption that the decomposition of all doctors is similar to that of hospital medical staff and Unrestricted Principals and Equivalents in general practice which together constitute 96 per cent of all medical doctors and dentists in the NHS, England. Source: OECD based on Department of Health Statistical Bulletin 2003/02, 2003/03 and 2003/04, June 2003.

2.4 practising physicians per 1 000 population (up from 1.9 in 1998), compared to currently 3.3 in the Netherlands and France. Looking ahead, several factors may add to the current workforce shortages. In particular, the implementation of the EU working-time directive will cause physician working hours to shrink as the maximum working week for junior doctors is to gradually decline towards 48 hours in 2009, some 6-10 hours below what many junior doctors work currently. Adding to this, many public sector employees will soon retire. And criticism is growing because of the problems created for countries like South Africa where shortages of doctors and nurses relative to the population’s medical needs are more pronounced than in the United Kingdom. These obstacles will not make it easier to meet the planned increase in the workforce requiring the number of medical doctors and dentists to grow a further 10½ per cent by 2008 relative to the September 2002 level. For nurses the required growth of a further 5 per cent relative to the September 2002 level is less challenging. The need to avoid unwarranted wage increases is all the more important as there is no apparent need to raise wages for structural reasons to attract personnel into these services over the longer term. Recruitment is rather a question of time and training more professionals since, at least in education, salaries are not low compared to countries with a similar standard of living (Figure 3.7). In England as well as Scotland the average pay for teachers in primary and secondary school is about 15-20 per cent higher than the average pay for all employees. They

© OECD 2004


79

Figure 3.7. Teacher salaries from an international perspective Index, average salary1 in whole economy = 100, 2001

200

200

180

180

160

160

140

140

120

120

100

100

80

80

60

60

40

40

20

20

0

0 DEU

JPN

CHE

AUS

IRE

ENG

SCO

FIN

NLD

USA

ITA

FRA

SWE

CZE

1. Teachers’ salaries are annual statutory salaries in public institutions at 15 years of experience as a weighted average for primary and secondary schools based on a standardized weight being the relative share of students at these levels for the total of OECD countries. Source: OECD.

are paid relatively more than in France, the United States and the Nordic countries, albeit relatively less than in Germany and Japan.12 Studies comparing public and private sector pay levels within the United Kingdom, while controlling for length of education and other individual characteristics, find that pay levels for men in the public sector are currently similar to those in the private sector while pay levels for women are higher in the public sector. This differential, however, has declined during the economic upturn in the second half of the 1990s (Disney and Gosling, 2003; Benito, 2000). … turning more input into more output… Judging by the available measures, increases in output lag the increases in input and have yet fully to materialise (Figure 3.8, panel A and B). While these measures of service output are far from ideal they are the best measures currently available.13 And comparing input and output measures, productivity seems to have declined as the growth in the number of doctors, nurses, hospital buildings and equipment have (not yet) been fully reflected in a growing number of treatments. In fact, growth in the volume of health care output has slowed down compared to the first half of the 1990s.14 Likewise, growth in the number of teachers and school buildings has not been fully reflected in growing numbers of students entering nursery school or staying on in post-compulsory secondary education. By bringing time series for spending, inputs and outputs together, panel A and B of

© OECD 2004


80

Figure 3.8. Performance in health care and education 1995 = 100

160

1995 = 100

160

160

150

150

140

140

130

130

120

120

120

120

110

110

110

110

100

100

100

100

90

90

90

90

80

80

200

100

180

80

80

160

160

60

60

140

140

40

40

120

120

20

100

0

A. Health

care 1

150 140

Expenditure in real terms Volume of input Volume of output: Outpatient treatments

130

80

1991

1995

2000

2003

B. Education

160

1,2

150 140

Expenditure in real terms Volume of input Volume of output

1991

1995

130

2000

2003

Per 100 000 population

200

100

Per cent

C. Death rates for persons under 75

180

Cancer Heart disease, stroke and related illnesses

1991

1995

Days, median waiting times

60

2000

2003

800

E. Waiting times

100

D. 11-year-olds achieving expected standards

20

English Mathematics

1991

1995

2000

2003

Thousand persons

0

Per cent

100

100

F. 15-year-olds’ achievements

700

55

600 50

500

80

80

60

60

40

40

400

45 40

Outpatient treatments (left scale) Inpatient, > 3 months (right scale) Inpatient, > 12 months (right scale)

35 30

80

300 200

20

1991

1995

2000

2003

0

0

20

Five C-grades or better Five G-grades or better

100 1991

1995

2000

2003

0

1. Input and output volumes indices reflect a broad set of physical measures (Pritchard, 2003). The broad output index for health care parallels an outpatient treatment which is therefore shown here given its longer time span. 2. Includes nursery, primary, secondary and special education, but excludes further education and universities. Source: Office for National Statistics, HM Treasury, Department of Health and OECD.

© OECD 2004


81

Figure 3.8 show how the origin of the gap between real spending and output growth differs between health care and education: in health care, the gap arises partly because the volume of input has grown more slowly than expenditure and partly because the extra inputs have not led to a fully commensurate increase in output, whereas in education the rise in spending has been fully converted into a rise in inputs. … and better outcomes Spending in excess of output growth may be warranted, however, if service quality improves and better outcomes are achieved. The goal of health care is a healthy population – not a large number of treatments – and for example prevention and better use of screening to detect diseases at an early stage can therefore eventually produce better value for money. However, in the health sector there are few indicators showing unambiguous improvements in outcomes over and above trend improvements that were already apparent before the surge in spending. Reductions in mortality from cancer and heart disease (Figure 3.8, panel C) is a case in point: while a comprehensive set of initiatives targeted at these major killers in the late 1990s may have contributed to the continued decline, it is difficult to see any break so far in the trend already established. Waiting times have started to come down recently and today nobody waits more than 12 months for hospital treatment, but at 50 days, the median waiting time for outpatient treatment is still above what it was when spending increases began (Figure 3.8, panel E).15 Quality as experienced by NHS patients may have improved, but reliable comparisons over time are scarce. A standardised patient survey of general practitioners (family doctors) found only negligible change from 1998 to 2002 with a stable and high share of patients having positive views of the general practitioner’s skills, knowledge, attitude, ability to communicate as well of the amount of time spent with the patient. However, patients reported that they more often had to wait for an appointment in 2002 than in 1998, despite the frequency of doctor visits having increased only slightly. For education, the measure of output volume used in the national accounts is simply the number of students taught and it does not reflect how much they learn.16 The simple volume measure should therefore be seen in a context where the share of 11-year-olds that achieve the expected standard of literacy and numeracy have gone up as documented by test results for English and mathematics (Figure 3.8, panel D). On top of this, geographic inequality in learning outcomes has been reduced as the number of schools where fewer than 65 per cent of students reach the expected standard has been halved since 1996. Improvements have been less marked in secondary education where the share of 15-yearolds achieving at least 5 passes at grade C or better has continued its slower trend rise, while the share achieving the less demanding grade G or better has flattened

© OECD 2004

82


(Figure 3.8, panel F). Considerable improvements in educational outcomes were achieved before education spending began to increase significantly through the introduction of pedagogic strategies and other reforms. Ultimately, the jury is still out on whether the massive spending increases will fully pay off in terms of improved service outcomes, when the extra resources have been absorbed and the new staff is fully productive. As for other large investments, productivity and quality pay-offs may take time to materialise. Nevertheless, the limited evidence of improvement in service outcomes beyond the trend already seen in the early 1990s, together with the large increases in costs and prices suggests that the expansion may have reached a speed limit and a more gradual increase in health and education spending could therefore improve the chance of locking in improved performance rather than higher costs. Performance management, targets and incentives An integral part of the government’s strategy for transforming increased expenditure on public services into better outcomes is a framework for performance management based on extensive outcome measurement and targets. National targets are set out in each government department’s Public Service Agreement which is based on and revised in the biennial Spending Reviews (Table 3.3). More detailed targets are specified within executive bodies such as the NHS and in the business plan of each service provider such as each hospital.17 Broadly speaking, such national targets for service delivery serve four purposes:18 – Support political transparencyand accountability. By giving a clear statement of what the government is trying to achieve, targets have been instrumental in gaining public support for more collectively financed spending. Moreover, the commitment to quantitative targets creates a strong political accountability as witnessed by the exceptional case when the Minister for Education resigned in 2002 because the target for literacy and numeracy of 11 year-olds was not fully met. – Build policies around desired outcomes, not just based on existing organisational structures. Examples are “Sure Start”, a cross-departmental programme aimed at supporting families and children before and after birth, as well as the 1998 target to reduce the death rate from accidents which led to the formation of a cross-government task force. – Concentrate effort on key challenges and raise ambitions for service delivery. Even if the original target for 11 year-olds’ literacy and numeracy was not fully met, many observers argue that having a specific quantitative target has helped to raise the ambition of teachers and headmasters and break the otherwise low expectations for children from disadvantaged backgrounds.

© OECD 2004


Table 3.3.

83

Examples of targets in the Public Service Agreements for health care and education

Target

Progress

Health care 1998 Spending review containing 24 targets in total • To reduce the death rates from: – cancer amongst people aged under 75 by at least 20 per cent by 2010; – heart disease and stroke amongst people aged under 75 by at least 40 per cent by 2010; – accidents by at least 20 per cent by 2010.

In early stage of delivery: 1998-2000 averages show: – a reduction of 6.3 per cent for cancer; – a reduction of 13.7 per cent for heart disease and stroke; – a slight increase of 0.5 per cent for accidents.

• Get doctors to prescribe generic drugs instead of the more expensive brand-name drugs having the same effect. Reduce by half or more the number of practices prescribing less than 40 per cent generics.

Met: The number of practices prescribing less than 40 per cent generics was reduced to 34 in 2002 compared to 598 originally.

• Reduce staff sickness absence in the Department of Health. Later, the target was specified as a reduction to 7.9 days a year by 2001 and 6.8 days a year by 2003.

A new electronic system for sickness reporting is now being piloted.

2000 Spending Review containing 10 targets in total • Narrow the health gap in childhood and throughout life differences between socioeconomic groups and between the most deprived areas and the rest of the country. Reduce the gap in infant mortality and life expectancy by 10 per cent by 2010.

Too early to assess.

• Reduce the maximum waiting time by end of 2005: – to 3 months for an outpatient appointment; – to 6 months for inpatient treatment.

On course: Status for England by September 2003: – 168 145 patients waiting longer than 3 months for an outpatient appointment; – 180 272 patients waiting longer than 6 months for inpatient treatment.

• Secure year-on-year improvements in hospital cleanliness and food.

On course/partly met: Various indicators show improvements and today no single hospital has very poor cleanliness (rated as “red”), compared to 35 per cent in 2000.

2002 Spending Review containing 12 targets in total • Further reduce the maximum waiting time by 2008 to 3 months for day cases and inpatient treatment. • Improve value for money in the NHS and personal social services by at least 2 per cent a year, with improvements of 1 per cent in both cost efficiency and service effectiveness.

© OECD 2004

Status for England by September 2003 is that 434 432 patients are waiting longer than 3 months for day cases and inpatient treatment. Indicators are still under development.


84

Table 3.3.

Examples of targets in the Public Service Agreements for health care and education (cont.)

Target

Progress

Education 1998 Spending review • Improve literacy and numeracy of 11-year-olds so that by 2002: – 80 per cent achieve the standard expected for their age (level 4) in English tests; – 75 per cent achieve the standard expected for their age (level 4) in mathematics tests. 2000 Spending Review containing 13 targets in total • Establish benchmarking by December 2002 so schools can compare costs with one another. 2002 Spending Review containing 11 targets in total • Improve literacy and numeracy of 11-year-olds further so that by 2004, 85 per cent achieve level 4. Reduce the number of schools significantly by 2006 in which fewer than 65 per cent of the 11-year-olds achieve this level.

Not fully met: The achievements of 11-year-olds has improved significantly, but by 2002 the target was not fully met as the expected standard was achieved by 75 per cent in English and 73 per cent in mathematics.

Met: A standardised framework for financial reporting is now in place.

See above.

• Reduce school truancy by 10 per cent by 2004 compared to 2002. • Set challenging targets for value for money in Further Education colleges.

Methodology under development.

Sure start 2002 Spending Review containing 4 targets in total • Reduce the number of 0-3-year-olds living in Postponement: This target was originally households where no-one is working by 12 per supposed to have been reached by 2004. cent by 2005-06. Source:

Departmental reports for the Depart of Health (1998, 1999, 2000, 2001, 2002, 2003b) and the Department for Education and Skills (1999; 2002; 2003).

– Keep the pace of activity and efficiency high in general. Used cautiously, targets can be useful in replicating the constant pressure on service providers to pursue efficiency that in other sectors of the economy arise from competition. Since its introduction in 1998, the framework has evolved in response to the teething problems that inevitably occur after any major innovation, and the degree of ambition of the various targets has been adjusted along the way. Delivering on some of the targets referred to in Table 3.3 has been postponed (such as the Sure Start target concerning small children living in families where

© OECD 2004


85

no one is working), some have been enhanced (such as the maximum waiting time target set in 2000, which was shortened in 2002), and consecutive spending reviews have moved towards fewer targets. With the Spending Review to be concluded by summer 2004, there will no longer be detailed Service Delivery Agreements at department level, but only Public Service Agreements leaving more room for local flexibility. Most of the several hundred targets set in 1998 for which it is possible to assess delivery havebeen met by now – indicating a high degree of success. However, a weakness of the current target framework is that some of the inherent problems with traditional public sector management have re-emerged. The 1998 Spending Review introduced a range of fairly detailed targets directing the operation of service provision, as the framework has evolved PSA targets have become fewer and more outcome focussed. A recent report by the Public Administration Select Committee of the House of Commons: “On target? Government by Measurement” (Box 3.3) concluded that the increase in accountability and transparency which targets have brought with them has been valuable, but also brought to the fore some examples of the practical problems associated with extensive performance measurement. One such example is of hospital trolleys being re-designated as “beds on wheels” and of corridors being re-designated as “pre-admission units” in order to circumvent waiting time targets for accident and emergency treatment. If not resting on very specific procedures and definitions, performance measurement can be surprisingly difficult and targets counterproductive. For the measurement of ambulance response times, unclear specification over when to start the clock and what to classify as a life threatening emergency undermined the basis of measurement, and because the clock can stop either when an ambulance or qualified responder arrives, many ambulance services have hired lay responders who may not be trained or equipped to handle the range of emergency conditions to which they are sent. The Select Committee also reported on the National Audit Office’s examination of a selection of department performance measurement systems and validated performance data: in over 80 per cent of such “first time” validations, it found that the organisation had materially misstated their achievements or had failed to disclose potential weaknesses with their data. The lesson to draw is that a large set of detailed targets does not guarantee operational efficiency. Given the transaction costs involved, a good target is one that serves several of the purposes listed above while being simple to measure and audit. In this sense, the targets for educational achievement are well chosen because they capture a fairly comprehensive measure of learning outcome without directing in detail how schools operate, while relying on an already existing system of tests keeping the additional measurement costs to a minimum. By contrast, the difficulties of measuring clinical outcomes are seen as a serious obstacle to the development of good performance measures in health care (Smith, Mannion et al., 2002).

© OECD 2004


86

Box 3.3.

Public Administration Select Committee report on performance targets

The Public Administration Select Committee of the House of Commons (2003a) recently published a report on public sector performance targets. Overall it concluded that it is worthwhile maintaining the target-based framework and that the increase in accountability and transparency which targets have brought with them has been valuable. However, it also had a number of key recommendations for improving it that are summarised below: 1. Greater local autonomy to construct more meaningful targets, which should be fewer and limited to the key outcomes. 2. Widening the target consultation process to involve professionals, service users, select committees and Parliament. 3. Strengthen the role of the National Audit Office (NAO) to not only audit the reporting systems, but also the performance measures. 4. Moving away from a “hit or miss” approach towards measures of progress. The government accepted many of the recommendations as being in line with the way it sought to evolve the current framework for the next spending review (Public Administration Select Committee, 2003b). However, given the external validation of departmental information systems the government, quite reasonably, did not see the need for external validation of all performance data, particularly given the additional administrative costs involved. In response to the recommendation for giving greater local autonomy in the setting of targets, the government stated that it would encourage departments to consult stakeholders more, that it endorses devolving decision-making from the centre, while maintaining national targets and service standards. On this last point, the extent and manner by which target setting reflects local conditions will perhaps be a key future development of the framework. Giving the local level a larger say on what should be their targets can promote a feeling of ownership and thereby commitment from professionals, and should therefore be piloted. On the other hand, if target setting is devolved to bodies where local providers have a large say, the result could be less ambitious targets and in the end more complacency with poorly performing providers. Maintaining pressure on managers such as school head masters with nuanced and realistic – but still ambitious – performance targets is therefore important. An important aspect rather neglected by the Select Committee is strengthening the incentives of service providers to ensure targets are met, for example through activity-based funding (as discussed further in the text).

Getting incentives right is crucial to achieve targets Even where central government has sufficient information to set relevant targets and measure outcomes, it still faces a basic implementation problem, because setting targets only affects behaviour by service providers, if they have

© OECD 2004


87

an incentive and feel motivated to meet the targets. The main challenge for the UK performance management framework is therefore how funding mechanisms, organisations and the role of consumers can be adjusted to craft incentives to improve services and thereby meet the key targets, while reducing the need for detailed national targets and measurement. The 1998 target on generic drug prescription illustrates that target setting, measurement and incentives can go hand in hand, but more can be done to craft incentives for service providers and users to change their behaviour. Pharmaceuticals currently account for about 15 per cent of UK health spending and, as in other OECD countries, this share has grown over the last decade. One way to contain spending is to urge physicians to prescribe generic drugs when there is no clinical difference compared with the more expensive brand-name drugs, and the 1998 Spending Review established a target for this.19 Currently, patients that are not exempt from charges pay a fixed £6.30 per item irrespective of whether a generic or brand-name drug is prescribed.20 Medical practitioners have some incentive to choose low-cost treatment as the price exceeding the fixed charge is carried by their Primary Care Trusts (the group of 50-200 medical practitioners to which they belong). Other OECD countries, including France, have introduced reference price systems setting the costs carried by the public purse at the price of a generic drug and letting the patient pay the difference if a more expensive drug with similar content and effect is chosen. Changing the way patients are charged in the United Kingdom in this direction would reinforce the incentives for making savings on pharmaceuticals without making access to the appropriate treatment dependent on ability to pay, as people with low income or chronic illness would continue to be exempt. Another option along these lines is to charge at least a symbolic amount from patients that do not show up for planned hospital consultations or surgery. Finally, broad exemptions from drug charges could be reconsidered as currently not only older persons with low income but all older persons are exempt from charges.21 Current health policy places greater emphasis on patient choice, performance incentives linked to the target framework and the transformation of NHS hospitals into Foundation Trusts. The first Foundation Hospitals will be in place by spring 2004 with some independence to borrow for investment and to depart from standard NHS employment conditions and pay levels. Granting public service providers more operational flexibility is an important element in health care reform in most OECD countries having public health services, but more freedom has to be complemented by stronger performance incentives. The main incentive to meet targets currently stems from the pressure on managers to perform, as their careers could be set back if the hospital lags significantly behind. Under these circumstances, benchmarking and target setting are effective by making it clear what is an unacceptable under-performance and by giving a clear deadline for service providers to improve.22 Good performers will be allowed to earn autonomy, and

© OECD 2004

88


thus a condition for getting foundation status is that hospitals have acquired three stars in the rating by the Commission for Health Improvement. The intention is that all hospitals will acquire foundation trust status within the next five years. League tables such as the hospital star ratings, however, have been criticised for being overly simplistic, and it is probably too early to assess the effectiveness of incentives to earn this autonomy. One-off sums or “cigar box money” have sometimes been used to promote a target. For example a one-off sum of £14 million to pay ward housekeepers was allocated to those hospitals that demonstrated commitment to achieve the 2000 target for hospital cleanliness and food. But in general, NHS budget allocation and funding mechanisms are not designed with a view to create incentives to raise activity to the same extent as in other OECD countries. Reforming the way budgets are allocated within the NHS, towards activitybased funding mechanisms is therefore likely to be a fruitful avenue and, the payment by results policy is a significant step in this direction. The experience from a number of OECD countries using activity-based funding for health care is that it raises activity, sometimes significantly, and thereby is a key factor for reducing waiting times (Box 3.4). Moreover, reducing long waiting times via funding incentives to raise activity rather than trying to enforce maximum waiting time guarantees by targets, would leave more room for hospitals and doctors to take local circumstances and patients’ individual needs into account. Some of the conflicts between maximum waiting time targets and clinical prioritisation frequently reported by physicians might then be avoided.23 The NHS began the phased introduction of activity-based funding for hospitals in April 2003. Funding will continue to flow from central government via Primary Care Trusts on a capitation basis, but whereas Primary Care Trusts today typically pay hospitals based on historic cost levels, they will increasingly allocate funding among hospitals based on activity. In their agreement with each hospital, Primary Care Trusts are to stipulate caps for the maximum activity they will fund, in order not to overspend their budget. The incentive for hospitals to raise activity thereby hinges on how the tariffs are set and adjusted over time, because if tariffs are too high, hospitals have only limited incentives to raise activity, as they quickly hit the cap on funding. To avoid this barrier, budget control could instead be ensured by adjusting tariffs automatically in response to total activity to match the budget envelope, as done in Austria. Or a part of the overall NHS budget could be reserved as additional (partial) funding where hospital activity increases beyond a benchmark. Experience from Denmark shows that this model can help raise activity and bring down waiting times (Clemmesen and Hansen, 2003). The plans to implement activity-based funding for hospitals should therefore be strongly endorsed. They will also support the wider choice for patients to be introduced in 2005, which can help address quality aspects that are difficult to quantify but easy to experience for users (Lundsgaard, 2003). Finally, the plans to involve a broader mix of public and private providers seems promising such as with the Fast Track Surgery Centres,

© OECD 2004


Box 3.4.

89

Activity-based funding, incentives and waiting times in health care

Making funding for hospitals and doctors’ individual pay depend on activity can bring an important contribution to tackling waiting times by creating incentives to use existing capacity better and treat more patients. A recent OECD study has compared countries with and without waiting times for elective surgery and found that countries are lesslikely to report problems with waiting times if they rely mainly on activity-based funding for hospitals rather than mainly fixed budgets and if they pay hospital doctors on a fee-for-service basis rather than a salary basis. The difference is statistically significant when controlling for other factors affecting supply and demand of health care including public and private spending, hospital capacity (number of beds), number of doctors and population age structure (Siciliani and Hurst, 2003). The finding is illustrated in Figure 3.9 which shows that higher spending helps to reduce waiting times as France, Germany, Switzerland and the United States which do not report waiting time problems all spend a large share of their GDP on health care. However, spending differences are not a sufficient explanation as for example Belgium without waiting times spends no more than Denmark, Netherlands and Australia all of which have substantial waiting time problems. What appears to be equally important is the incentives provided by funding mechanisms and none of the four countries that had mainly activity-based funding in 2000 reported waiting time problems – despite the fact that Austria and Japan had comparatively low levels of spending. A correlation of average waiting times with doctors’ pay being either mainly salary, mixed or mainly fee-for-service gives similar results to that in Figure 3.9. For the NHS to achieve a durable reduction in waiting times, increased funding as well asimproved incentives would be necessary. Other policy levers to reduce waiting times include clinical guidelines for prioritizing patients as applied in New Zealand (Hurst and Siciliani, 2003).

where a handful of foreign firms are to provide elective surgery such as hip replacements on contract with the NHS. While competition can in some instances be problematic in health care, it is likely to work well in the case of such standard treatments.24 In parallel with activity-based funding for hospitals, rewarding extra work via individual pay could help compensate for the shortage of doctors and counteract the decline in average hours worked by doctors. Today, doctors employed at NHS hospitals are paid a fixed salary, and only if treating private patients after normal hours will their income depend on activity. The consequence is rather adverse incentives for doctors to avoid reductions in waiting times in order not to lose the extra income from high-paying private patients wishing to avoid the queue. In this context, paying hospital doctors a mixture of salary and fees as for example in Austria, Switzerland and with managed care in the United States,

© OECD 2004


90

Figure 3.9. Waiting times, spending and incentives 2000 Average waiting time, days 250

1

1

Average waiting time, days 250

Finland Finland

England

England

200

200

Australia Australia

150

Spain

150

Norway Spain

Norway

100

100

0

6

8 10 12 14 Total health expenditure, % of GDP

Netherlands

Austria Japan United States 3 France 2

France

United States

Germany Switzerland

France

Belgium

Austria Japan

Luxembourg

50

Denmark

Belgium Germany Luxembourg Switzerland United States 3

Netherlands

Denmark

50

0 Mainly fixed budgets

Mixed

Mainly activitybased funding

1. Mean waiting times for persons admitted for inpatient surgery. Simple average for hip replacement, knee replacement, cataract surgery, varicose veins, cholecystectomy, and inguinal and femoral hernia. 2. In France, public hospitals have fixed budgets, while private hospitals treating publicly funded patients receive activity-based funding. 3. In the United States, Health Maintenance Organisations use mixed funding mechanisms, while the public Medicare programme uses activity-based funding. Source: OECD Health Data 2003 and Siciliani and Hurst (2003).

would seem to not only help motivate additional effort, but could also balance the adverse incentive to keep waiting times long. Unfortunately, the three year agreement with hospital doctors finally accepted in late October does not include standard elements of incentive pay, and therefore a key recommendation is to move ahead with piloting and implementation of incentive pay for hospital doctors faster and more broadly than currently planned as this would help reduce waiting times (Box 3.4). In primary health care, a new contract for general practitioners will relate a substantial share of pay to quality and outcome measures supplementing the traditional capitation based payment.25 The new contract implies that general practitioners will have less incentive to refer to hospitals in cases where it is possible to treat the patient in primary care, and this will help realising the cost savings

© OECD 2004


91

associated with shifting from in-patient to ambulatory treatments that are made possible by advances in medical technology including pharmaceuticals. Long-term public finances In a welcome move to enhance the transparency and credibility of fiscal policy the government recently published an update of what are intended to be annual reports on the longer term outlook for public finances, including some (inevitably tentative) fiscal projections to the middle of the century (H.M. Treasury, 2003c). These projections tend to confirm a conclusion from wider comparative exercises across many OECD countries, namely that the United Kingdom is one of the few OECD countries that on current policies faces smaller long-term fiscal pressures from population ageing (OECD, 2001). A recent independent study (Cardarelli et al., 1999), which developed a first set of generational accounts for the United Kingdom, concluded that unlike the United States, Japan, Germany or Italy, the United Kingdom, under an assumed fiscally responsible baseline (intended to characterise the current policy stance) “did not exhibit a large imbalance that seriously threatens the well-being of future generations”, although generational balance was still not achieved. The smaller fiscal pressures are only in part due to the fact that population ageing is less rapid than in much of the rest of the European Union or Japan (but not the United States), with the UK old age dependency ratio still expected to rise significantly between the beginning and mid-century. A more important factor is that the public pension system is less generous than in most other OECD countries and one of the system’s key components, the basic state pension, is indexed to changes in prices rather than earnings. In this respect, as discussed further in Chapter VI, the government are tackling pensioner poverty by the use of targeted income related benefits, and measures to bolster the already well developed occupational pensions system along with measures to promote other forms of retirement saving. The major pressures on ageing are instead more likely to come from the need for higher health expenditure which in the Treasury report are estimated to rise by a further 1½ per cent of GDP between 2007-08 (the final year of the current spending plans for the NHS) and the middle of the century. Assessment Both the OECD and 2003 PBR projections suggest that the government’s golden rule will be met over the current cycle, although in both cases the safety margin is not large. However, an important difference between the two projections is that whereas the official projections suggest the fiscal position will be steadily strengthening and be on track to meet the golden rule over the next cycle, there is little improvement in either the current or overall financial deficit according to the OECD projections, suggesting the need for further fiscal action. Even on this more

© OECD 2004

92


pessimistic view, the fiscal position, both in the long term and in terms of government debt as a proportion of GDP, would remain stronger than in most other major OECD countries. However, in this case, failure to take action would risk diminishing the credibility of the governments’ fiscal framework. Moreover, adjustment, if needed, would be better started during the current upswing rather than being postponed with the risk of reinforcing the downswing of the next cycle. Firm cost control in non-priority spending areas is therefore advisable and much of the required adjustment could be achieved by slowing the rate of increase in public expenditure. But independently of macroeconomic considerations, there is a case for re-profiling the planned expenditure increases in certain areas of health and education to ensure optimal systemic readiness and capacity, with more gradual spending growth overall. Increased expenditure must deliver better outcomes over and above trend improvements that were visible before the spending increase was initiated. Raising spending in priority areas may be necessary to reach satisfactory service quantity and quality, but the speed with which the increase is taking place is generating cost pressures within the public sector. Extending the period over which increases take place would also help the process of piloting promising innovations – such as giving hospitals more operational flexibility, activity-based funding mechanisms, incentive pay for hospital doctors and Fast Track Surgeries – and learning from experience before implementing large spending increases. More operational flexibility in terms of human resource management is key to improved efficiency in health care where traditions and barriers between staff categories inhibit the introduction of modern flexible work practices. The flipside of the coin is that this inevitably increases competition for human resources even with national wage agreements, and reforms such as the introduction of Foundation Hospitals are therefore likely to provide the biggest pay-off in a context of more gradual spending growth. Firm control of cost pressures while targeting the additional investment in health and education spending will increase the chance of locking in improved performance rather than higher costs. The deteriorating fiscal position also highlights a basic strategic choice that the United Kingdom faces in improving health care to a level common in many parts of continental Europe as well as further raising standards in education: either to accept the corresponding level of taxes and mandatory contributions that are typical in Continental Europe, or to expand private funding and provision in ways that do not compromise equity concerns. If substantial tax rises are to be avoided, users will have to contribute part of the costs of improving health care and higher and continued education. The obvious case for this is in higher education, where the private returns are probably the highest in any OECD country because universities are mostly publicly funded despite a wide earnings differential between persons with and without higher education. There is a clear case for

© OECD 2004


93

developing ways of allowing students to channel more of their future income into universities, thereby reducing the burden on public finances (Chapter IV). In health care, the government clearly wants to maintain and strengthen the principles underlying the NHS, namely that access to treatment does not depend on ability to pay, but within that framework there is room for extending cost-sharing, as currently the share of total health care spending coming from private sources is among the lowest in any OECD country. As discussed above, charging differently for non-generic drugs or charging for missed appointments, can improve incentives at the same time as raising revenue. Getting the health care system right and finding the appropriate level of spending is all the more important because the most powerful fiscal pressures to the middle of the current century as a result of ageing comes from health care rather than pensions in the case of the United Kingdom. Avoiding locking in inefficiencies as a result of too-fast spending growth is therefore also crucial in a longer term fiscal perspective.

© OECD 2004


94

Notes 1. Eschenbach and Schuknecht (2003) estimate that the sensitivity of company taxes and taxes on financial and capital transactions in the United Kingdom are among the highest of 17 OECD countries examined. 2. This is based on the logarithmic change in the FTSE all share index from 3008 to 2040 between fiscal year 2000 and fiscal year 2002. 3. Eschenbach and Schuknecht (2003) estimate that for a 10 per cent fall in equity prices direct taxes on companies fall by 0.13 per cent of GDP and taxes on financial and capital transactions by 0.04 per cent of GDP. The 2002 Pre-Budget Report suggests that a 23 per cent fall in equity prices reduced tax receipts from capital taxes, stamp duty and, because of their impact on the capital gains of life insurance companies, corporation tax, by about £3½ billion (0.4 per cent of GDP) after one year. 4. These projections apply to the general government only, whereas the government applies its fiscal rules, including the golden rule to the public sector as whole. However, this should only marginally affect the figures as the difference between general government and public sector borrowing as a per cent of GDP is usually small. 5. The “central” assumption is of potential growth of 2¾ per cent per annum to fiscal year 2006-07, declining to 2½ per cent thereafter. For details of this calculation see H.M. Treasury (2002c). 6. In the last two years of the projection period, in fiscal years 2007 and 2008, the Treasury lowers the trend growth rate used to project the public finances to 2¼ per cent, as a consequence of estimated demographic trends. 7. The Eurocare-3 study compared five-year survival for persons diagnosed with cancer during 1990-94 and followed to the end of 1999. Average survival rates for men in England, Scotland and Wales as well as Denmark was in the 33-37 per cent interval compared to 40-47 per cent in most Nordic and Western European countries (Eurocare-3, 2003). A similar difference was found for women. See also BMJ Editorial (1999). 8. A Eurobarometer poll from spring 1998 found that 43 per cent were very dissatisfied or fairly dissatisfied. This, however, was similar to the opinions expressed in other EU countries (EC, 1998). More recently, a poll among sicker adults from 2002 found that only 31 per cent expressed themselves as not at all satisfied or not very satisfied, which was fewer than in Australia, Canada, New Zealand and the United States. Also, patients less frequently reported problems with conflicting information from different doctors and medical mistakes in the United Kingdom than in the other four countries (Blendon et al., 2003). 9. The decision was announced on 4 December 2003, and a preliminary report will be produced by July 2004 followed by a final report by January 2005. See Box A4 in H.M. Treasury (2003b) for details.

© OECD 2004


95

10. Nevertheless, the rapid growth in the use of nurses from temporary employment agencies indicates that some parts of the NHS find it difficult to recruit nurses. Already in 2001-02, expenditure on agency nursing had reached a level corresponding to 7-8 per cent of the total salary bill for nurses in NHS hospitals, more than twice its 1997-98 level. To counteract this development, all NHS hospitals will be required to source temporary staff exclusively via a new in-house agency, NHS Professionals, by April 2005. 11. As an example, the new NHS pay system to be implemented in October 2004 has supplements for nurses and midwives in high-cost areas giving 15 per cent and 20 per cent extra pay in outer and inner London respectively, and 5 per cent extra pay in Fringe Zones. The new supplements represent an increase compared to the current supplement levels of £3 333 a year in Inner London, £2 604 a year in Outer London, and £729 a year in Fringe Zones. 12. On average across OECD countries, the salary of a full time teacher in upper secondary school is 42 per cent higher than that of a full time teacher in primary school, while in the UK the difference is more limited. That is, for primary school, UK teachers are better paid compared to their colleagues in other countries, while the opposite holds for upper secondary school. 13. The Office for National Statistics is working towards more comprehensive measures of government output and productivity (Pritchard, 2003), but currently the best available datasets are based on simple physical measures of output such as the number of schoolchildren taught. 14. The growth in outpatient attendances in Figure 3.8 refers to so-called first out-patient attendances, that is the number of cases being treated. This grew by a total of 13.0 per cent from 1997-98 to 2002-03. For comparison, the total number of outpatient attendances grew by only 7.1 per cent over the same period reflecting that the average patient now attends a hospital fewer times during an outpatient treatment. This reflects either improved productivity or less thorough follow-ups as there are some indications that post-treatment controls are being cancelled unduly as hospitals are forced to meet waiting times targets (Public Administration Select Committee, 2003a). Hospital admissions have increased by an average of 1.9 per cent annually from 1999 to 2002, i.e. since the spending increase started. This is less than the 2.9 per cent annual increase from 1991 to 1999, before spending rose. The limited increase in not made up for by a rise in the number of outpatients. The number of outpatients treated has increased by an average of only 2.4 per cent annually from 1999 to 2002 compared to a 3.9 per cent annual increase from 1991 to 1999 (Figure 7.1 in Deparment of Health, 2003b). Changing the composition of health care away from hospitalization towards less costly outpatient treatment could be an avenue for improving value-for-money, but as shown by these numbers, there has not been a rise in the number of outpatients that could make up for the limited rise in hospital admissions over recent years. Finally, it should be mentioned that the combined output volume index for health care calculated by the Office for National Statistics is roughly similar to the volume of outpatient treatment shown (Pritchard, 2003). 15. The total wait for a patient needing surgery is typically a combination of the waiting times shown in Figure 3.8. For example at Moorfields Eye Hospital, average waits are in the range of 2-3 months for a first consultation and thereafter another 2-5 months before the surgery takes place. www.moorfields.nhs.uk/AboutUs/WaitingTimes (accessed 19 October 2003).

© OECD 2004

96


16. Hence the government’s objective to limit class sizes reduces the national accounts measure of productivity in education. 17. To give an example, the 2002-03 business plan for Moorfields Eye Hospital in London (which is one of the NHS hospitals applying to become a Foundation Trust next year) contains 65 targets. Some targets are quantitative measures (such as to reduce nursing vacancies to 8 per cent) while others call for the hospital to address a specific problem (such as by developing a strategy for screening) or to implement new procedures (such as to copy clinical correspondence to patients). The business plan replicates the national waiting time targets but is in some cases more ambitious. Source: www.moorfields.org.uk/ (accessed 19 October 2003). 18. In the UK government’s official publications it is typically stated that targets can give a clear statement of what the government is trying to achieve, a clear sense of direction and ambition, a focus on delivering results, a basis for judging what is and is not working, and better accountability. 19. Generic drugs can appear when a patent expires and other producers can copy a particular drug. The original producer may often be able to charge a higher price because of its well known brand name and hence enjoy an economic rent. Consequently, getting physicians to prescribe the generic drug can bring down costs, in some cases considerably. 20. It is also possible to purchase a pre-payment certificate for £32.90 for four months, and holders of this certificate are not charged per item. 21. Another illustration of the issues of balancing measurement and targets versus incentives is the way the problem of high sickness absence in the Department of Health was addressed in 1998. The quantitative target listed in Table 3.3 presumably helped to raise awareness about the problem, but the fact that it has taken a new computer system which is still today in a pilot phase illustrates the difficulty and transaction costs associated with measurement. The alternative would be to alter the way budgets are allocated among divisions of the department to give them a stronger incentive to economize on resources internally and in doing so reduce sickness absence. With stronger incentives in place, the need for centralized monitoring of the exact extent of sickness absence would be limited. Also employees could be given an incentive to keep sickness absence low, either by not paying salary for the first day of absence or by giving a bonus to those not being on sick leave for extended periods. 22. The apparent importance of managerial career concerns is an important aspect to note for other countries considering adopting targets for performance management in publicly funded services. The effectiveness of targets depends crucially on the rules and culture surrounding job security for managers in public services, and the UK framework might therefore be much less effective if implemented in other European countries. 23. One example of this was given by a hospital director to the Public Administration Select Committee explaining that over-ambitious targets for waiting times reductions has led to systematic cancellation of follow-up appointments after eye surgery with the consequence that some patients lost vision. 24. Involving a broader mix of providers can stimulate productivity as public and private providers learn from each others innovations, and consequently a recent study found for the United States, that public expenditure were lower in areas where some hospitals are run as private firms (McClellan and Kessler, 2002). 25. Capitation based payment gives general practitioners a fixed payment per payment on their “list” with adjustments for factors such as age and gender.

© OECD 2004


97

Bibliography Benito, A. (2000), Public Sector Wage Differentials in the 1990s, Bank of England. BMJ Editorial (1999), “Cancer Survival in Britain”, British Medical Journal, Vol. 319. Blendon, R.J., C. Schoen, C. DesRoches, R. Osborn, and K. Zapert (2003), “Common Concerns Amid Diverse Systems: Health Care Experiences in Five Countries”, Health Affairs, Vol. 22. Carderelli, R., J. Sefton, and L.J. Kotlikoff (1999), “Generational Accounting in the UK”, National Institute for Economic and Social Research Discussion Paper, available at www.niesr.ac.uk/ pubs/dps/dp147.pdf Clemmesen, F. and M. Hansen (2003), “Erfaringer med meraktivitetsfinansiering af sygehuse”, Samfundsøkonomen, No. 3. Department for Education and Skills (1999), Departmental Report, The Government’s Expenditure Plans 1999-00 to 2001-02. Department for Education and Skills (2002), Departmental Report 2002. Department for Education and Skills (2003), Departmental Report 2003. Department of Health (1998), The Departmental Report 1998-99. Department of Health (1999), The Departmental Report 1999-2000. Department of Health (2000), The Departmental Report 2000-01. Department of Health (2001), The Departmental Report 2001-02. Department of Health (2002a), The Departmental Report 2002-02. Department of Health (2002b), Reforming NHS Financial Flows: Introducing Payment by Results, October 2002. www.doh.gov.uk/nhsfinancialreforms/financialflowsoct02.htm. Department of Health (2003a), Delivering the HR in The NHS Plan 2003. Department of Health (2003b), Departmental Report 2003-04. Disney, R. and A. Gosling (2003), A New Method for Estimating Public Sector Pay Premia: Evidence from Britain in the 1990s, CEPR Discussion Paper, No. 3787, February. Eschenbach, F., and L. Schuknecht, (2003), “The Fiscal Costs of Financial Instability Revisited”, ECB Working Paper, No. 191. Eurocare-3 (2003), “Cancer Survival in Europe at the End of the 20th Century”, Annals of Oncology, forthcoming. EC (European Commission) (1998), “Europeans, Health, and the Healthcare System”, Eurobarometer Special Survey. Glennerster, H. (2002), “United Kingdom Education 1997-2001”, Oxford Review of Economic Policy, Vol. 18, No. 2.

© OECD 2004

98


Giorno, C, P. Richardson, D. Roseveare, and P. Van Den Noord (1995), “Estimating Potential Output, Output Gaps and Structural Budget Balances”, OECD Economics Department Working Papers, No. 152. H.M. Treasury, (1998), “Code for Fiscal Stability”, www.hm-treasury.gov.uk. H.M. Treasury, (1999), “Fiscal Policy: Public Finances and the Cycle”, March, www.hm-treasury.gov.uk/. H.M. Treasury, (2002a), “End of Year Fiscal Report”, November, www.hm-treasury.gov.uk/. H.M. Treasury (2002b), Cross-Cutting Review of the Public Sector Labour Market, November 2002, www.hm-treasury.gov.uk/. H.M. Treasury (2002c), “Trend Growth: Recent Developments and Prospects”, April, www.hmtreasury.gov.uk/. H.M. Treasury (2003a), Budget 2003, www.hm-treasury.gov.uk/. H.M. Treasury (2003b), Pre-Budget Report, December 2003, www.hm-treasury.gov.uk/. H.M. Treasury, (2003c), “Long-term Public Finance Report: Fiscal Sustainability of an Aging Population”, December, www.hm-treasury.gov.uk/. Hurst, J. and L. Siciliani (2003), “Tackling Excessive Waiting Times for Elective Surgery: A Comparison of Policies in Twelve OECD Countries”, OECD Health Working Papers, No. 6. Jonsson, B. and U.G. Gerdtham (2000), “Health Care Systems Internationally Compared” in A.J. Culyer and J.P. Newhouse (eds.), Handbook of Health Economics, Vol. 1A., Elsevier. Lundsgaard, J. (2003), Competition and Efficiency in Publicly Funded Services„, OECD Economic Studies, No. 35, OECD, Paris. McClellan, M. and D. Kessler (2002), “The Effect of Hospital Ownership on Medical Productivity”, RAND Journal of Economics, Vol. 33, No. 3. NIESR (National Institute for Economic and Social Research) (2003), National Institute Economic Review, November. OECD (2001), “Fiscal Implications of Ageing; Projections of Age-Related Spending”, OECD Economic Outlook, June. OECD (2003), Health at a Glance, OECD, Paris. Pritchard, A. (2003), “Understanding Government Output and Productivity”, Economic Trends, No. 596, Office for National Statistics, July. Public Administration Select Committee (2003a), On Target? Government by Measurement, Fifth report of the Session 2002-3, HC 62-I, published on 22 July 2003, London. www.publications.parliament.uk/pa/cm200203/cmselect/cmpubadm/62/6202.htm. Public Administration Select Committee (2003b), On Target? Government By Measurement: the Government’s Response to the Committee’s Fifth Report, Sixth report of the Session 2002-3, HC 1264, published on 11 November 2003, London. www.publications.parliament.uk/pa/cm200203/cmselect/cmpubadm/62/6202.htm. Siciliani, L. and J. Hurst (2003), Explaining Waiting Times Variation for Elective Surgery across OECD Countries, OECD Health Working Papers, No. 7. Smith, P., R. Mannion and M. Goddard (2002), “Performance Management in Health Care: Information, Incentives and Culture” in Public Services Productivity, paper presented at a seminar held at H.M. Treasury on 13 June 2002.

© OECD 2004


99

Annex 3.A1

An equation for corporation tax receipts This annex reports an estimated equation explaining (non-North Sea) corporation tax receipts as a share of GDP (CTSH) in terms of the OECD’s measure of the output gap (GAP), the Financial Times Stock Exchange all-share Index (FTSE) index normalised on money GDP and the corporation tax rate (TR). The equation is estimated on annual data over a sample period from 1980 to 2002. ∆CTSH = 9.105 + 0.3098 ∆CTSH–1 – 0.1008 ∆GAP + 0.0903 ∆GAP–1 (1.1) (1.3) (1.8) (2.3) +1.2350 ∆InFTSE + 0.3695 ∆InFTSE–1 – 0.5581CTSH–1 + 0.0503 ∆TR–1 (2.0) (0.9) (–2.7) (1.6) Rsqd-adjusted = 0.607, t-statistics reported in brackets, Standard error of regression = 0.246, LM – test for up to second order serial correlation = 1.04 (p-value = 0.38), Ramsey RESET test = 0.79 (p-value = 0.47), Jarque-Bera test for normality of residuals = 0.44 (p-value = 0.80), White heteroskedasticity test = 0.91 (p-value = 0.58). The long-run responses suggest that: – A 1 percentage point closing of the output gap (i.e. actual output rising in relation to potential output) will raise the share of corporation tax receipts in GDP by 0.16 percentage point. – A 10 per cent increase in the FTSE-all-share index (relative to money GDP) will raise the share of corporation tax receipts in GDP by 0.07 percentage point, although the long-run response is not well determined. – No long-run effect from the corporation tax rate was found to be significant although the equation does include a lagged short-run dynamic effect.

© OECD 2004


100

Annex 3.A2.

Potential output growth estimates

This annex describes the current OECD estimates of potential output growth and compares them with those of the Treasury. The OECD method of calculating potential output utilises a production function approach based on a two factor, capital and labour, Cobb-Douglas production function (Giorno et al., 1995). The Treasury methodology instead combines estimates of trend labour productivity and trend labour input, where the latter is broken down into components of average hours, the trend employment rate and the population of working age (H.M. Treasury, 2002c). The OECD projection of potential output The OECD projects potential output growth over the period 2002 to 2006 of 2.5 per cent per annum, down from the 2.7 per cent per annum over the period 1997 to 2002 (Table 3.A2.1). The decline in potential growth relative to the recent past can be attributed to two factors: – Firstly, the trend employment rate is expected to increase by less than in the recent past. This is mainly because the structural unemployment rate is expected to fall 0.2 percentage point between 2002 and 2006 (an average of 0.05 percentage point per annum), compared with a fall of 0.2 percentage point on average each year between 1997 and 2002. The trend participation rate (on working-age basis) is expected to be flat between 2002 and 2006, in contrast with an 0.1 percentage point increase between 1997 and 2002. – Secondly, trend output per hour is also expected to decline from 2.1 to 1.9 per cent per annum. This is, nevertheless, consistent with a stable growth contribution of multifactor productivity, because over the recent past the capital stock grew faster than both output and labour input so that there was a temporary “capital deepening” effect which boosted labour productivity. With the growth of the capital stock expected to be slower over the next few years the capital deepening effect will be reduced and lead to slightly lower labour productivity growth. Two factors which partially offset these negative effects on overall potential growth are: – Trend average hours worked are expected to fall, but less markedly than in the recent past given that some of the recent reduction may be attributed to one-off factors such as compliance with the EU working time directive. – The population of working age is projected to grow by over 0.5 per cent per annum, reflecting official demographic projections by the Government Actuary Department (GAD), but adjusted to have higher net immigration of 150 000 per annum (rather than

© OECD 2004


Table 3.A2.1.

101

Decomposition of trend growth

Annual averge growth per cent Trend output per hour

Trend average hours

OECD estimates/ projections 1997-2002 2002-06

2.07 1.86

–0.23 –0.07

0.35 0.11

Treasury estimates/ projections 1997 H1 2001 Q3 2001 Q4 to 2006 Q4

1.96 2.00

–0.37 –0.10

0.36 0.20

Trend Population employment of working rate age

Total trend output1

Multi-factor productivity2

Capital stock2

0.49 0.54

2.67 2.45

1.11 1.14

4.35 3.57

0.66 0.60

2.60 2.75

n.a. n.a.

n.a. n.a.

1. Sum of first four columns. 2. Figures not given for the Treasury numbers because they are not relevant to the methodology they use. Source: OECD and H.M. Treasury (2002c).

100 000 per annum assumed by the GAD), given very recent upward data revisions to immigration statistics. Comparison with the Treasury’s potential output projection The Treasury estimate that potential output will grow by 2¾ per cent per annum between 2001 Q4 and 2006 Q4, compared with 2.9 per cent trend output growth over the recent past. It should, however, be noted that the Treasury use a “deliberately cautious” estimate of potential growth for the public finance projections. Thus given the central estimate of 2¾ per cent per annum in Table 3.A2.1, a lower potential growth assumption of 2½ per cent per annum is adopted for the public finance projections which the NAO deemed “reasonable and cautious” in its Budget 2002 audit. Despite the differences in the methodologies used and the resultant estimates, both the OECD and the Treasury forecasts are based on the assumption that the productivity growth going forward will be ¼ of a percentage point lower than over the recent past. However the OECD estimates imply that trend productivity growth will be 1.9 per cent per annum going forward whereas the Treasury estimates it to be 2¼ per cent per annum. Examining the other components of potential growth there is one minor difference with the OECD calculations. The Treasury has a slightly larger (approximately 0.1 percentage point) contribution from trend employment growth than the OECD, with the difference explained by a somewhat higher contribution from the labour force participation rate (see above). With the exceptions of trend productivity and employment, both estimates are based on similar growth assumptions with respect to the trend average hours and population of working age components of potential growth. Other estimates The National Institute for Economic and Social Research estimate that potential output growth will average 2.4 per cent per annum between 2002 and 2006 (NIESR, 2003). These calculations are based on the chain-linked rebased national accounts data. In the concluding statement for the Article IV Consultation (December 2003), the International Monetary Fund refers to a trend rate of output growth “slightly above 2½ per cent”.

© OECD 2004

IV.

Policies to enhance potential growth

Introduction While macroeconomic policy has been very successful in ensuring that output remains close to potential, the evidence is less compelling that structural policy has succeeded in lifting potential output growth closer to best performance.1 Productivity gaps with the best OECD performers remain large, as discussed in Chapter I, despite a wide range of policies addressing this weakness. While employment rates are relatively high, they fall well short of the highest in the OECD, and the proportion of the population that is not engaged in the labour market at all has remained flat over recent decades, so that the major source of increased labour input has come through a reduction in structural unemployment. The remainder of this Chapter assesses what areas are most promising in promoting potential growth, considering in turn, labour utilisation and productivity. A summary of progress on structural reforms, following-up and updating recommendations from the previous Survey, can be found in Annex A. Labour utilisation The UK unemployment rate of around 5 per cent will be the lowest among the major seven in 2003. Inactivity rates are about 7 percentage points below the EU average, but are similar to those in Canada, the United States and Australia. They are, however, clearly higher than in a small group of mainly Nordic countries.2 Moreover, while the structural unemployment rate has fallen by around 4 percentage points since 1990 there has been virtually no fall in the trend inactivity rate (Figure 4.1). The flat aggregate inactivity rate conceals a number of worrying trends. While the female inactivity rate has fallen,3 the male inactivity rate has shown a consistent upward trend. The latter has been accompanied by a similar rise in men reporting long-term sickness or disability as the main reason for inactivity, which as a proportion of the inactive population (excluding students), has risen from about 40 per cent in the mid-1980s to a peak of nearly 60 per cent in the late 1990s, from which it has only declined slightly (Jones et al., 2003). There has also been a trebling of the numbers on invalidity benefit during the 1980s and the

© OECD 2004


104

Figure 4.1. Unemployment and inactivity rates Per cent

Per cent

30

30

25

25 Inactivity rate 2 Trend inactivity rate 2

20

20

Unemployment rate 1 NAIRU 1

15

15

10

10

5

5

0

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

0

1. Per cent of labour force. 2. Per cent of working-age population. Source: OECD.

first half of the 1990s, and since 1995 the number claiming incapacity benefit (which replaced invalidity benefit) has continued to rise, albeit at a slower rate (Figure 4.2).4 In 1980 the numbers claiming invalidity benefit were less than the number claiming unemployment benefit, whereas they are now more than twoand-half times as great. The increased numbers on incapacity benefit is partly explained by an increase in the average duration of benefits: nearly half of those on incapacity benefit have been receiving them for over five years (compared to 5 per cent for unemployment benefit). Once on incapacity benefit for a year claimants only have a one in five chance of returning to work within five years. This rise in incapacity has been much more pronounced among those with low skill levels which together with the strong correlation between unemployment rates and the incidence of disability-related benefits at the regional level, suggests that economic factors are driving these trends. Indeed, in some localities within regions the concentration of disability claimants has reached alarming proportions.5 There is also evidence that higher inactivity rates among older men with lower qualifications are occurring at successively earlier cohorts: men in the 193539 birth cohort with no qualifications had an inactivity rate of 8 per cent when aged 45-49 compared with a rate of 26 per cent for those of the same age born between 1950 and 1954 (Barham, 2002a). Nevertheless, the share of the low skilled in total inactivity has declined, reflecting the general rise in educational attainment in the population, while there

© OECD 2004


Figure 4.2.

105

Working age claimants of incapacity-related benefits and unemployment Thousands

3500

3500

3000

3000

2500

2500

2000

2000

1500

1500

1000

1000

500

500

Claimants of incapacity-related benefits Unemployment claimant count

0

1980

1982

1984

1986

1988

1990

1992

1994

1996

1998

2000

2002

0

Source: Department for Work and Pension and OECD.

has been a marked tendency for the inactive population to become increasingly male and older (Jones et al., 2003). The inactivity of older workers due to illness or disability is among the highest in the OECD (Figure 4.3), and, in common with most other countries where this is the case, may partly reflect disability benefits being used as an alternative pathway to early retirement given the absence of other “formal” early retirement schemes. This is, however, far from a complete explanation because the UK ranking among all OECD countries in terms of inactivity rates due to disability is even higher among lower age groups (OECD, 2003a), with, for example, nearly two-thirds of inactive men aged 35-49 giving their reason for inactivity as being long-term sick or disabled in 2001 (Barham, 2002b). It seems rather that part of the reason for the increase in inactivity is because disability schemes may have acted as a “pressure-valve” as access to other benefits, at all ages, has become tighter. Another concern has been the high incidence of workless households, particularly those with children (Gregg and Wadsworth, 1999; 2003). In the mid1990s more than one in five working age households (defined as those with at least one adult aged 15-64) had no member of the family in employment, which was among the highest in the OECD (OECD, 2002a). Since then there has been a fall in the incidence of workless households by around 3 percentage points, to a level which is more typical of other OECD countries. The non-employment rate

© OECD 2004


106

Figure 4.3. Inactivity of older workers due to illness or disability Persons aged 50-64 years, 2000 Per cent of population

Per cent of population

25

25

20

20

15

15

10

10

5

5

0

BEL

SVK

KOR

GRC

AUS

DEU

ITA

CHE

HUN

ISL

OECD1

PRT

ESP

LUX

CZE

NLD

DNK

UK

FIN

0 POL

1. Population-weighted average for countries shown. Source: OECD (2003b), Employment Outlook, Paris.

among working age households with children has also fallen, but especially for lone parents remains high relative to other OECD countries: in 2001 the nonemployment rate in lone parent households was around 50 per cent compared to an average of just over 30 per cent among 15 OECD countries (OECD, 2002a). Policies to reduce unemployment and inactivity One of the main measures to address the issue of workless households has been the Working Families Tax Credit (WFTC) introduced in 1999 to provide in-work financial support for families with children. A number of studies, summarised in Blundell and Reed (2000), suggest that this measure had a positive impact on the employment in families previously without earners, in particular raising participation by around 60 000. Nevertheless, around half of this impact is likely to have been offset by a reduction in the employment rate of married women with a working partner.6 Thus while the WFTC may have been successful in terms of reducing the incidence of workless households, its net effect on raising overall labour supply is probably modest. In April 2003 the government effectively combined several existing parts of the tax and benefit system that supported families, including the WFTC, and replaced them with the Child Tax Credit and the Working Tax Credit. In early October 2003 the latter was providing in-work support for around 1.7 million single people and couples, with and without children, in low

© OECD 2004


107

paid work – the first such measure to help the low paid without children. A preliminary independent evaluation (Brewer, 2003) suggests that while the change may have a significant impact on raising the income of families with children and the working poor, as for the WFTC, work incentives improved for some and worsened for others. The United Kingdom spends relatively little on active labour market measures, around 0.4 per cent of GDP in 2001-02, compared to most OECD countries, although this partly reflects the relatively low level of unemployment as the ratio of active to passive labour market measures (unemployment benefits) is relatively high. The most important active labour market programmes designed to directly reduce unemployment and inactivity are the welfare-to-work programmes under the umbrella of the “New Deal,” which give special attention to disadvantaged groups (see Box 4.1), the largest of which cover young people, the long-term unemployed (those aged over 25) and lone parents (Table 4.1). The proportion of New Deal participants finding a job on leaving the programme varies considerably across the schemes. For the long-term unemployed just over one-fifth go into jobs that last more than 13 weeks. While this appears quite low it should be remembered that in the context of a relatively tight labour market those who have been unemployed for 18 months are likely to have the lowest levels of skills. Through participation in the New Deal it is possible that participants gain work experience and new skills even if they do not get a lasting job immediately. For young people leaving the NDYP the proportion finding a job that lasts more than 13 weeks rises to about one-third. Evaluations of the NDYP which allow for substitution/displacement effects and for the impact on equilibrium wages, suggest that unemployed young men are about 20 per cent more likely to get jobs as a result of the policy, so raising youth employment by about 17 000, with the social benefits of the policy exceeding the costs. Much of the employment effect is estimated to result from the wage subsidy element, but at least one-fifth is attributed to an enhanced job search (Van Reenen, 2001; Blundell et al., 2002). These results appear encouraging as many evaluation studies of youth labour market measures across a number of OECD countries have reported little impact (Martin and Grubb, 2001). Official evaluations of the New Deal 50 plus while apparently positive (Department for Work and Pensions, 2003a), have been mainly based on qualitative data and a small number of interviews. The voluntary nature of the scheme (in contrast to the other main New Deal programmes) may explain the high job retention rates as a result of self-selection, but also suggests large deadweight losses. The take-up of the training grant also remains disappointingly low. Independent assessments have concluded that the increase in the employment rate of the

© OECD 2004


108

Box 4.1.

New Deal Programmes

The New Deal for young people (NDYP), introduced in April 1998, is mandatory for 18- to 24-year olds who have been claiming unemployment benefit for 6 months. Participants first enter a “Gateway” period lasting up to 4 months during which advisors work with them to improve their employability and to find unsubsidized jobs for as many as possible. Those who do not find a job move on to one of four options (where figures in brackets indicate the proportion on each option as of June 2003): a period of subsidized employment (12 per cent); a course of fulltime education or training (45 per cent); a job with an environmental task force (19 per cent); or a job in the voluntary sector (23 per cent). On reaching the end of this option without obtaining a job the participant will normally claim unemployment benefit and will enter a follow-through period during which time they receive intensive help to find a job. The New Deal for long-term unemployed, introduced in April 1998,is mandatory for those aged 25 and overthat have been claiming unemployment benefit for more than 18 months. Participants first enter a “Gateway” period lasting up to four months followed by an “Intensive Activity Period” (mandatory for those aged 25-49, voluntary for those over 50) where various options are available including work experience, work placements with employers, occupational training, and help with motivation and soft skills. It is possible for participants to move into subsidized employment at any stage of the programme. Those who return to unemployment benefit at the end of this period will undertake a period, usually six weeks, of intensive job search designed to build on the experience of the earlier part of the programme. The New Deal 50 plus, introduced nationally in 2000, is targeted at those aged between 50 and state pension age and is a voluntary programme which helps both unemployed and economically inactive people and their dependent partners, who have been claiming benefits for six months or more, into paid work. The aim of the programme is to help people aged 50+ back into employment by personal advice and help with job search. If the participant finds employment with an income of less than £15 000 they receive a tax free employment credit of £60 per week for a fulltime post and of £40 per week for a part-time post, for up to a year, although this has recently been merged with, and paid through, the Working Tax Credit. In addition a training grant can be paid to those receiving an employment credit. The New Deal for lone parents (NDLP) provides a comprehensive package of support, including access to a personal adviser; help with training, education and childcare; and advice on benefits, in-work financial support and self-employment. Eligibility for NDLP has been extended to all lone parents who are either not working or who work less than 16 hours a week. The New Deal for partners (NDP) is a personal adviser service for the partners of benefit claimants, extending work-focused help beyond the main claimant in a household for the first time. It was introduced in 1999 and will be enhanced to provide the same package of help and support as NDLP in April 2004, coinciding with the introduction of compulsory work-focused interviews for the partners of benefit claimants.

© OECD 2004


109

Box 4.1. New Deal Programmes (cont.) The New Deal for disabled people (NDDP) was extended across England, Scotland and Wales in July 2001 to build on pilot experiences. The NDDP is the first national programme specifically designed to help those on incapacity-related benefits and provides a national network of job brokers to help disabled people locate and move into secure employment. Participation in the NDDP is voluntary.

Table 4.1.

New Deal summary statistics1 Long-term unemployed

New Deal for lone parents

Disabled people

1 045 970 89 420

651 0002 58 270

540 630 98 790

57 1763 n.a.

462 870

164 8602

254 230

20 350

98 000

n.a. 80

n.a. 30

n.a. 10

Young people

Number of participants since start of programme Currently participating Number of jobs to date found by leavers Percentage of jobs lasting more than 13 weeks Expenditure 2002-03 (£ million)

79 260

78 210

Over 50s

n.a. n.a.

1. As at September 2003, except for New Deal for over 50s which is for March 2003. 2. New Deal for Long term Unemployed people aged 25+: including both pre April 2001entarants and entrants from April 2001. 3. Registrations. Source: Department for Work and Pensions and H.M. Treasury (2003a).

target age-group was more a consequence of favourable demand conditions (Disney and Hawkes, 2003). For lone parents the proportion finding a job on leaving the programme, at 52 per cent, is considerably higher than with other programmes. The success of this programme together with the WFTC has been a major factor in raising the employment rate of lone parents from just over 40 per cent throughout the early 1990s to over 53 per cent in 2003 and in increasing the number of hours in work among those already in employment (Gregg and Harkness, 2003). The success of the NDLP programme is confirmed by a cost-benefit study which suggests that there is a net economic gain of £4 400 per additional job from the NDLP programme with a net saving to the exchequer of £1 600 per job (Department for Work and Pensions, 2003b). The government has an ambitious target to further raise the employment rate of lone parents to 70 per cent by 2010, although this target looks challenging.

© OECD 2004

110


The much smaller scale of the New Deal for the disabled reflects its more recent start, and the voluntary nature of participation. Viewed in the context of the increase in inactivity due to sickness and disability over the last couple of decades, the scale of policy response so far appears inadequate. Around threequarters of new claimants have more manageable medical conditions such as back pain or depression, which should not represent an insurmountable barrier to work given adequate support.7 The United Kingdom is one of few OECD countries that has an overall employment rate which is above the OECD average, but for which the employment rate for disabled people is below the average (OECD, 2002a). Moreover, while a disabled person entering the labour market for the first time may need additional vocational training, in 1999 the United Kingdom had only rather few vocational rehabilitation schemes (normalised on disability benefit inflows) and a very low per capita spending on vocational rehabilitation (normalised on disability benefit costs) among a range of OECD countries (OECD, 2003b). The UK response has been to initiate a comprehensive framework for reform building on the potential for work of people on incapacity benefits and shifting the focus away from their incapacity. The aim is to adopt a more holistic approach that addresses all the potential barriers to a return to work, encompassing not only health issues but also financial issues, lack of skills and of vocational rehabilitation, and the need for expert guidance and support. Starting from October 2003 this radically new approach to dealing with those on incapacity benefits is being piloted in seven areas with an intensive series of mandatory workfocused interviews for new claimants of incapacity benefit and access to new and improved employment programme provision, including condition management programmes, developed jointly with local health service providers. The pilots will also test new financial incentives, including a “Return to Work” credit to be paid for the first year for those moving from incapacity benefit into work (Department for Work and Pensions, 2002). The credit will be paid at £40 per week for one year where personal income in work is less than £15 000 a year, and should significantly improve the gains from entering low-paid, and especially part-time, employment. New claimants with severe conditions who are exempt from the mandatory requirements and existing recipients will have voluntary access to the pilot provisions. The pilot scheme will be evaluated thoroughly before any decision on extending it nationally is taken. Closing the productivity gap Closing the productivity gap with its major competitors has been a major theme motivating UK structural reforms. One of the key drivers of productivity, namely competition is considered in the next chapter. The focus of this Chapter is on areas where the United Kingdom has had long-standing deficiencies relative to its main competitors, namely in skills and investment.

© OECD 2004


111

Raising skills The recent OECD Growth Project highlighted the importance of human capital in explaining differences in growth performance among OECD countries. In this respect the United Kingdom compares poorly on a range of indicators with its major competitors, suggesting this has been a major factor holding back productivity performance. Applying the results from the Growth Project, namely that a 10 per cent increase in the stock of human capital would increase per-capita GDP by between 4 and 7 per cent (OECD, 2003a), to measures of the human capital stock across countries, suggest that if the UK working age population had the same level of human capital as in Germany, GDP per capita could be 5-10 per cent higher (Figure 4.4). And relative to Australia, Canada and the United States, differences in human capital can also explain much of the considerable productivity gap. While educational attainment of students has improved in recent years and the number of people having academic qualifications is relatively high, the current UK workforce appears to be less skilled than in many other OECD countries both in terms of basic literacy and vocational skills. Nearly one quarter of the adult population lacks basic literacy skills meaning that they may, for example, be unable to determine the correct amount of medicine to take from information printed on the package (level 1 in Figure 4.5). This is more than twice as many as Figure 4.4. Human capital explains part of the productivity gap Estimated effect on UK GDP per capita if UK human capital was at par with that of other OECD countries,1 1998 Per cent

Per cent

15

15

♦

10

10

♦

5

♦

♦

0 -5

5 0

♦

-5

♦ Maximum estimate

-10

♦

-10

Minimum estimate

-15

-15 France

Germany

Italy

Australia

Canada

United States

1. Based on the finding in the OECD Growth Study that a 10 per cent increase in the average number of years of education of the working-age population produces an increase in GDP per capita of 4-7 per cent (OECD, 2003c). Source: OECD.

© OECD 2004


112

Figure 4.5. Basic literacy of the adult population and of the young1

80

Percentage of 16-65-year-olds at different literacy levels, 1996

2

80

40

40

0

0

40

40

80

80

Level 1 SWE

DNK

Level 2 FIN

Level 3 DEU

CZE

Level 4/5 CAN

AUS

USA

Percentage of 15-year-olds at different literacy levels, 2000

UK

NZD

IRE

80

3

80

40

40

0

0

40

40

80

Level 1 and below FIN

CAN

Level 2 IRE

NZD

Level 3 AUS

UK

SWE

Level 4/5 USA

DNK

CZE

DEU

80

1. The two figures are based on different tests and therefore the levels cannot be compared directly. 2. The figure shows so-called document literacy, that is, the ability to locate and use information contained in various formats. The international ranking of the United Kingdom is similar for prose literacy (understand and use information from texts) and quantitative literacy (apply arithmetic operations). 3. The figure shows the combined reading literacy, that is, the ability to retrieve information, interpret texts, reflect and evaluate. The international ranking of the United Kingdom is similar for mathematical literacy, but a bit better for science literacy. Source: OECD: International Adult Literacy Survey (IALS) and Programme for International Student Assessment (PISA).

in Germany, the Netherlands and the Scandinavian countries, and also more than in Australia and Canada.8 Moreover, relatively few have an apprenticeship, skilled craft or technician level qualification, and the skill shortfall is particularly pronounced for the younger cohorts on the labour market when compared to other countries. For the 25-34-year olds, one-third have little or no formal qualifications beyond compulsory education, a considerably larger share than in any other major OECD country (Figure 4.6).9 Reaping the full benefits of globalisation and new technologies requires the whole workforce to be endowed with a stronger skill base, and research suggests that for Information and Communication Technology (ICT) investments to be fully productive, changing organisational structures and a complementary investment

© OECD 2004


Figure 4.6.

Low1

113

Educational attainment of the adult population 2001

Upper secondary1

Tertiary with a practical, technical or occupational orientation

Tertiary with a theoretical orientation

A. Population aged 25-34 years

B. Population aged 45-54 years

JPN

USA

CHE

CHE

SWE

DEU

CAN

JPN

USA

CAN

FIN

SWE

DEU

FIN

FRA

UK

NLD

NLD

IRL

FRA

AUS

AUS

UK

IRL

0

20

40

60

80

100

0

20

40

60

80

100

1. “Low” comprises persons having primary school, lower secondary school or ISCED 3C short programmes as their only formal qualification; upper secondary includes also post-secondary non-tertiary programmes. Source: OECD, Education at a Glance 2003.

in human capital may be necessary. For example, comparing the automotive industry with that in Germany, Mason and Wagner (2002) find that the shortfall of investment in skills as well as equipment makes UK manufacturers less productive. The effect of these trends is highlighted by the growing wage differential between skilled and un-skilled and the dramatic rise in inactivity among the less skilled discussed earlier in this chapter. In this context, continuously raising average educational attainment should be a policy priority and it is particularly important to deal with the large number of people lacking the basic skills and competencies needed in a modern workplace. For the young generation studying and graduating now, the skills gap relative to comparable OECD countries is less pronounced. Indeed, literacy of British 15-year olds is among the best in terms of reading proficiency as measured by the OECD PISA study (Figure 4.5). But after compulsory education, a smaller share of 15-19-year olds is enrolled in secondary education than in many OECD countries (Table 4.2). Conversely, the share of 20-29-year olds studying and graduating with a degree from tertiary education is similar to or above that in comparable OECD countries. The educational trend with many persons receiving an academic education but fewer getting an apprenticeship, skilled craft or technician level qualification is being perpetuated for those entering the labour market now.10

© OECD 2004


114

Table 4.2.

Educational attainment of the young studying and graduating now 2001, per cent

Secondary education

Tertiary education

Share of 20-29-year olds studying1

Share of 15-19-year olds studying1

Germany Czech Republic France Sweden Netherlands Finland Switzerland Australia Ireland United States Canada United Kingdom Italy

89.4 87.8 86.6 86.4 86.2 85.3 83.3 81.1 80.9 77.6 75.0 74.7 72.2

Finland Sweden Australia Netherlands Germany United Kingdom United States Canada Switzerland France Italy Ireland Czech Republic

39.2 33.0 28.3 24.3 24.2 23.3 22.6 21.2 19.7 19.6 17.1 16.6 14.7

Graduates with a practical, technical or occupational orientation2

Graduates with a theoretical orientation2

7.3 4.0

40.7 29.6 42.0

10.7 11.5

19.0 37.4

16.1 17.9 0.3 19.0 5.0

18.7 25.0 20.0 29.3 14.1

1. The figures represent the total number of full-time and part-time students in the age group as a share of the total population in the age group. A high share can reflect that a large share is in education or that study times are long. 2. Number of graduates as a share of the total population at the typical age of graduation. Source: OECD, Education at a Glance 2003.

Much has been achieved regarding quality in compulsory education, as discussed in the Chapter III, as well as regarding universal provision of nursery education. These are the most important foundations for enabling all to fully participate in the labour market and to develop their human capital, skills and potential to their best. Against this background, policy attention should now be on: – Increasing the number of young people getting apprenticeship, skilled craft or technician level qualifications. A set of initiatives already drives in this direction, including efforts to strengthen links with industry in education programmes for the 14-19-years olds, expansion of modern apprenticeships to include one-in-four at age 22, and introduction of two-year foundation degrees in higher education with a vocational orientation. The challenge is for these vocational pathways to ensure a high quality reputation raising their value on the labour market and attracting those young people who leave school at an early age today.11 – Energizing the university sector and providing more funding via a graduate contribution scheme. British universities are generally of high quality, but seem to have lost ground over recent decades. Reforming the some-

© OECD 2004


115

times archaic university management systems as suggested by the Lambert Review of business-university collaboration (H.M. Treasury, 2003b) and making more funding available via the planned graduate contribution scheme (discussed below) is therefore essential. – Improving training of the existing workforce.Employers in the United Kingdom already fund more adult training than in many OECD countries (Table 4.3), but in light of the low average skill and literacy level, the Government’s strategy is to strengthen training further. Upgrading the skills of the existing workforce is valuable from both a personal and a productivity perspective, but only if learning outcomes are sufficient to warrant the costs of tuition, the loss of production from being away from work and the administrative burden for firms. Therefore, policy measures that provide flexibility such as programmes targeted at adults who keep working part time while undergoing training are welcome, and cost-sharing should be developed to ensure that relevant course content is chosen (Box 4.2).

Table 4.3.

Continued adult training and education

Average number of hours per year per person employed, 19961 Employer-sponsored

Denmark Netherlands United Kingdom Finland Australia United States Canada Czech Republic Ireland Switzerland Italy

Self-sponsored

Government-sponsored

Total

Training2

Formal education3

Total

Training2

Formal education3

Total

Training2

Formal education3

39 30 30 23 2 22 21 18 14 9 8

36 21 22 23 15 18 17 13 9 9 8

2 8 8 1 7 3 3 5 6 1 0

23 25 22 33 33 23 29 12 7 11 11

19 17 17 22 15 14 17 8 3 8 10

4 8 5 11 18 8 12 4 4 3 1

27 17 20 29 23 17 20 10 12 7 11

22 13 16 22 14 12 15 7 7 6 11

6 4 4 7 9 5 4 2 6 1 0

Note: Some training and education is sponsored from several sources, and therefore the totals cannot be added up. Countries are sorted by the extent of employer-sponsored training and education. 1. Data refer to job and career-related training and education of adults aged 26-65 years in 1994 for Canada, Ireland, the Netherlands, Switzerland (for German- and French-speaking regions) and the United States, in 1996 for Australia and the United Kingdom and in 1998 for Czech Republic, Denmark, Finland, Italy and Switzerland (for Italian-speaking regions). 2. Courses corresponding to an apprenticeship certificate; a professional or career upgrading, and other continued vocational training courses. 3. Courses corresponding to a university degree, diploma, certificate; a college diploma, certificate; a trade vocational diploma, certificate and elementary or secondary school diploma. Source: International Adult Literacy Survey (OECD and Statistics Canada, 2000).

© OECD 2004

116


Box 4.2. Upgrading the skills of adults In July 2003, the UK Government presented a Skills Strategy to make available training more responsive to the specific skill needs of employers and workers, and to raise the learning ambitions and demand for skills. Firms are to get more control over the publicly-funded training they receive and management, particularly in smaller firms, are to be supported by advice and training in their key role as human resource managers. The Employer Training Pilots started in September 2002 and recently extended with a third year to August 2005 and expanded to cover one-third of England. They are an important element of the strategy. In these pilots, employers that offer their low-skilled staff time off to train for basic qualifications get compensation for wage costs (can vary with size of firm and other arrangements) and training is free or heavily subsidised. Currently 20 000 learners or 0.07 per cent of the labour force are involved. Support for higher level vocational skills will also be enhanced in areas of sectoral or regional skill priorities. Dialogue with employers and labour unions are to play a bigger role in particular through Sectoral Skills Councils, and unions are expected to encourage the low skilled to engage in training. The qualifications framework is to be redefined with Modern Apprenticeships open also to adults, vocational routes to focus more on employability and enterprise skills and with learning programmes divided into smaller units making them more flexible. To reinforce the demand-led approach, further education colleges and other training providers are to be funded differently and private providers to be accredited for public funding. These measures are in addition to the target of taking 750 000 adults through basic literacy training of whom 400 000 have by now completed the course. Any strategy for raising adult skills has to balance social and economic growth objectives. Social policy concerns might call for focusing training on low-skilled persons irrespective of age and covering a broad array of course topics, but as a policy to enhance potential growth, training persons in relatively more skilled jobs to adapt to technological change may matter more, and how long an individual has left before retiring is in any case a key determinant for the economic returns of training. For an adult already in gainful employment, leaving this in order to return to school is only beneficial in cases where the training is adequate and the person strongly motivated. In particular, the cost of taking adults out of work should be borne in mind for the courses with a cultural, leisure, community and personal fulfilment purpose proposed in the Skills Strategy. Such courses should not be confused with the kind of up-skilling needed to reduce the productivity gap. In light of experience from other OECD countries, efforts to improve adult skills are likely to be more cost-effective the better they can address the market imperfections involved: – Better sharing of benefits between employers and workers. Without requiring public spending, regulatory provisions can support investments in training such as by allowing pay-back clauses specifying that a worker leaving a firm within a specified period after going through employer-funded training shall reimburse part of the cost incurred by the employer. Employers and unions play a key role in making these mechanisms work, and the dialogue with these parties is therefore a positive element of the skills strategy.

© OECD 2004


Box 4.2.

117

Upgrading the skills of adults (cont.)

– Make training manageable in daily life and overcome credit constraints of workers. Some workers are less likely to be offered training by employers (women, immigrants, young employees, involuntary part-time and temporary workers, workers in low-skilled occupations and workers with low literacy). Nurturing a supply of training for these groups that is compatible with daily life and within the economic capacity of families is essential. In Australian technical and further education colleges, the options of studying part-time, at distance and on weekends have become very popular and allow families to fund training by foregoing leisure and reducing consumption without the need to raise credit. Individual Learning Accounts exist in some countries and were introduced in the United Kingdom in 1998 but then abolished in 2002 following cases of abuse (York Consulting, 2002). – Public subsidies to employers and workers are second best options. Tax incentives or grants to firms are used in some OECD countries (e.g. Austria, France, Italy and the Netherlands), but dead-weight losses can be considerable. The Nordic countries have experience with heavily subsidised training for workers as well as unemployed. But paying close-to-normal income during a training leave (Sweden) or using training as a forced “activation” of the unemployed (Denmark) can imply that the person feels little motivation to learn and may adversely raise the reservation wage of the unemployed (Westergaard-Nielsen, 2001). In Sweden, the Adult Education Initiative introduced in 1997 has been downsized considerably recently because going through upper secondary school did not improve employment prospects of adults much (Stenberg, 2003) – even though it was quite expensive. Against this background, the intention with the UK skills strategy to give free tuition for basic level 2 qualifications in order to help all workers to a minimum skill level is reasonable, but the grants to full-time students being piloted should be seen as a second best compared to better availability of part-time, distance and weekend studies, allowing families to fund training by foregoing leisure and reducing consumption. Compensating employers for the wage cost when low-skilled staff attend training during working hours under the Employer Training Pilots helps to address the pressing issue of skill shortages, but it significantly increases the cost for the public purse. Source:

OECD (2002b), OECD (2003d), OECD (2003e), Ok and Tergeist (2003), and Department for Education and Skills (2003a).

Increasing funding for higher education via income-contingent graduate contributions is the fairest option Improving higher education is vital for economic growth, and whereas the United Kingdom and most European countries spend around 1 per cent of GDP on

© OECD 2004


118

higher education, Canada and the United States spend around 2½ per cent of GDP. The total level of spending on tertiary education tends to be higher in countries that rely not only on government spending but also have substantial contributions from students, donations and other non-government sources (Figure 4.7). Those countries – Canada, the United States, New Zealand and Korea – that have been able to channel more than 2 per cent of GDP into tertiary education all raise a substantial share of funding from these alternative sources. For the United Kingdom, at slightly over one-quarter, this share is higher than in many other OECD countries, but is well below the share in the countries that spend the most on tertiary education. Given the growing constraints on public finances, it would be difficult to raise large amounts of extra funding for British universities via general taxation, nor would it be fair when considering that the individuals endowed with education enjoy large gains from it. In particular for higher education, the private returns are large, and in the United Kingdom those with a university degree earn, on average, about twice as much as those without post-compulsory education. Because of this Figure 4.7.

Expenditure on tertiary education institutions 2000

Total expenditure as a percentage of GDP 3.0

3.0 USA KOR

CAN

2.5

2.0 FIN

SWE

DNK NOR

NLD

AUT DEU GRC

0

AUS

IRE BEL

1.0

0.5

2.0

NZD 1

1.5

2.5

ISL

FRA

MEX

1.5

ESP JPN

UK

1.0

ITA

10

20

30

40

50

60

70

0.5 80

Contributions from students, donations and other non-government sources as a share of total expenditure, per cent 2

1. For New Zealand, OECD education data only has information about public spending. Other spending components are estimated based on the Statement of Financial Performance for tertiary education institutions from the cash flow from tuition fees, revenue from services provided, investment income and other sources. Using the same data source to estimate public spending produces results that are consistent with what the New Zealand authorities report to the OECD Education data. 2. Net of tuition fees paid by government. Source: OECD, Education Database and Statement of Financial Performance for tertiary education institutions in New Zealand.

© OECD 2004


119

large wage differential and because, as in other European countries, most of the spending on higher education is paid out of the public purse, the average economic return to individuals taking higher education is probably the largest in the OECD (Blöndal et al., 2002). Against this background, the government’s plan for a graduate contribution scheme is commendable. Since 1998, students in all universities and across different subjects have paid an annual tuition fee, which was £1 125 in the academic year 2003/04. In practice, though, many are exempt because of the income situation of their family, with 43 per cent paying nothing and 16 per cent paying only part of the tuition fee. The problem with this arrangement is that if tuition fees are raised in order to meet the funding needs of universities, having to pay more up-front can become a barrier for students. The plan for a graduate contribution scheme launched in a recent white paper on “The future of higher education” (Department for Education and Skills, 2003b) will deal with this problem by allowing all students to defer the payment of all fees until they have graduated. The scheme will have two effects: it overcomes the credit constraint that would typically prevent students from borrowing against their future income; and it provides insurance as graduates will only be required to pay instalments when their annual income exceeds £15 000. Universities will be allowed to charge up to £3 000 a year, and a number of universities have indicated that they will go to this maximum level when the scheme is implemented from 2006. The exemptions based on family background will continue to apply for the basic fee of £1 125, and a new grant to students from low-income families will supplement the existing student loan for living costs which is available at a maximum £3 905 a year to students from lowincome families studying away from home.12 This package of changes will increase the opportunities for students from both high and low-income background, and the graduate contribution scheme will generate an amount equivalent to about 0.2 per cent of GDP.13 In the current UK debate, it is sometimes argued that funding higher education by individual loans rather than taxes could discourage students from poor families, and that it is unfair that very successful graduates will end up paying a smaller share of their income than less successful graduates. These arguments miss the point that social gradients in access to higher education, and equity in educational attainment more generally, are primarily determined by cognitive developments in early childhood and the foundation laid during school (Carneiro and Heckman, 2003). Indeed, much of the rise in education spending since 1997 has gone into improvements in nursery education with the Sure Start programme and in compulsory education notably improving the achievements by relatively more in schools in disadvantaged areas. Maintaining these improvements while expanding higher education based on contributions from those who benefit from it rather than based on general tax revenues is the most direct way to ensure equity in education outcomes. Introducing the graduate contribution scheme will

© OECD 2004

120


in a sense increase redistribution by channelling funding to students (that have low income while studying) and taking a contribution when they graduate and earn a high income (Goodman and Kaplan, 2003). Such redistribution over each person’s lifetime is preferable to fee exemptions, below-market interest rates and other subsidies for students with a low-income background which have ambiguous effects on equity, as they redistribute income to individuals that may be poor today, but over their lifetime are more affluent than the average tax payer (Barr, 2003a, 2003b). Rather than such subsidies, resources are better spent on other forms of pro-access policies, for example rewarding universities for attracting students from non-traditional backgrounds, as also proposed in the government’s white paper. Concerns have been voiced that it might become more difficult to recruit certain public professions like teachers and nurses that have fairly long education relative to their pay levels, as young people would opt for careers with better earnings opportunities. However, public employers could offer to cover part of their graduate contribution. A few other OECD countries have arrangements that are similar to the UK graduate contribution scheme (Annex 4.A1). In Sweden, as in the United Kingdom, student loans for living costs are repaid on an income-contingent basis, and large tuition fees are common in North America and Japan, but so far only a few countries including Australia and New Zealand fund tuition costs via income-contingent contributions from graduates. Consequently, the UK graduate contribution scheme could be a role model for other countries in Europe that may also have to reconsider the adequacy of their higher education systems in a modern knowledge-driven economy and vis-à-vis North America.14 Boosting investment Business investment has long been identified as one of the key drivers of growth, both through a transitory capital deepening effect and by embodying technical progress. The cross-country correlation between physical investment and productivity is both striking and robust (OECD, 2003c). While previous OECD EconomicSurveys have highlighted the historically low level of investment as a major factor holding back productivity growth in the United Kingdom, there are encouraging signs of an improvement. Real business investment grew rapidly over the second half of the 1990s, although it has been more sluggish since 2000 possibly partly related to the need to finance pension fund gaps (Chapter I). Nevertheless, in terms of business investment per worker, the gap with other major competitors has narrowed (Figure 4.8), although it is still significant vis-à-vis Australia, the United States, Canada and France. Macroeconomic stability over the last decade has no doubt contributed to this.

© OECD 2004


121

Figure 4.8. Business investment per worker US dollars, at 1995 prices and purchasing power parities, averages

10000

1980-89

1990-95

1996-2002

10000

8000

8000

6000

6000

4000

4000

2000

2000

0

0 Australia

France

Canada

United States

Italy

United Kingdom

Germany

Source: OECD.

Also encouraging has been the pick-up in ICT investment, with the United Kingdom being one of only half a dozen countries – which include the United States, Canada and Australia – where ICT investment now accounts for more than one-fifth of non-residential investment. This in turn reflects a finding that investment in ICT has been higher in those OECD countries with a more competitive domestic environment and which have undergone extensive product market reforms (Figure 4.9, upper panel). These countries have also found it easier to translate such investment into productivity improvements in ICT-using sectors, thereby increasing its contribution to aggregate productivity growth (Figure 4.9, lower panel). The real payoff, however, from higher ICT investment ought to come from higher multi-factor productivity growth. In this respect the United Kingdom – in contrast to the United States, Canada and Australia – has yet to see any improvement (Figure 4.10), a finding which is confirmed across a number of studies using different methodologies and data adjustments (Basu et al., 2003; Gust and Marquez, 2002). Basu et al. (2003.) in a comparison of the experience of the United States and the United Kingdom emphasise the role of ICT as a “general purpose technology” which may require reorganisation and learning and so only raise multi-factor productivity in ICT-using sectors with a long lag. They suggest, albeit tentatively, that the United Kingdom may therefore see an acceleration in multi-factor productivity growth over the next decade. However, a less optimistic interpretation is

© OECD 2004


122

Figure 4.9.

Regulation and investment in information and communication technology

ICT, 2001 or the latest available year 1

ICT, 2001 or the latest available year 1

30

30

28

28

Correlation coefficient: -0.60 t-statistic: -2.95

United States

26

26

24

24 United Kingdom

22

22

Canada Australia

20

20

18

18 Germany

16

Italy

14

16 14

France

12 10 0.0

12 0.5

1.0

1.5

2.0

2.5

10 3.5

3.0

Product market regulation (inward-oriented) 2,1998 ICT using services 3, 1996-2001

ICT using services 3, 1996-2001

1.6

1.6 Correlation coefficient: -0.62 t-statistic: -3.35

1.4

1.4

United States

1.2

1.2 Australia

1.0

1.0

United Kingdom

0.8

0.8

0.6

0.6 Canada

0.4

0.4

0.2

0.2

Germany

Italy

0.0

0.0 France

-0.2 0.0

0.5

1.0

1.5

2.0

2.5

-0.2 3.5

3.0

Product market regulation (inward-oriented) 2,1998

1. ICT investment as a percentage of non-residential gross fixed capital formation. 2. Indicator of economy-wide regulation excluding barriers to international trade and investment. The indicator ranges from 0 to 6, from least to most restrictive. 3. Contribution to aggregate labour productivity growth. Source: OECD.

© OECD 2004


Figure 4.10.

123

Pick-up in multi-factor productivity growth and increase in ICT investment

Change in ICT investment as % of GFCF, 1990-2000

14 12

14 Correlation coefficient: 0.58 t-statistic: 2.56

10

12 10

United Kingdom United States

8

Canada Australia

6

8 6

France

4

4 Germany

2 0 -1.5

2

Italy

-1.0

-0.5

0.0

0.5

1.0

0 1.5

Change in MFP growth from 1980-1990 to 1990-2000, per cent 1

1. Multi-factor productivity is defined as the residual term in a GDP per capita decomposition after accounting for the contributions of the labour and capital inputs. Source: OECD.

that other structural weaknesses have hampered an improvement in multi-factor productivity, the most obvious being the relatively low level of skills. At the industry level there is also a striking difference between the United States and United Kingdom in the “wholesaling and retailing” sector (important because of its size in the economy), with productivity in these sectors improving sharply in the United States, but at a slower rate in the United Kingdom. Another possible explanation for this difference is the greater scarcity of land in the United Kingdom combined with tight planning restrictions, discussed in detail in Chapter V, which may have reduced competition and so inhibited investment and productivity performance in this sector. The UK’s historical under-investment in the private sector has not been particularly apparent in plant and machinery investment (when expressed as a proportion of GDP), the category of investment generally thought of as embodying recent technical progress, but rather in non-residential construction investment and housing investment (IMF, 2003). Low levels of non-residential construction investment, in common with low levels of housing investment, as discussed in Chapter II, probably reflect overly rigid planning restrictions.15 Low levels of government investment (Figure 4.11) are a reflection of decades in which cuts were viewed as the easy option for constraining public expenditure. The government’s current fiscal rules, which make a distinction between current and capital expenditure (Chapter III), are a useful discipline to avoid such short-term expediency. Also

© OECD 2004


124

Figure 4.11. Government investment As a per cent of GDP

4.0

4.0 1980-89

3.5

1990-95

1996-2002

3.5

3.0

3.0

2.5

2.5

2.0

2.0

1.5

1.5

1.0

1.0

0.5

0.5

0.0

0.0 United States

France

Italy

Canada

Australia

Germany

United Kingdom

Source: OECD.

welcome is its objective to raise public net investment to around 2 per cent of GDP, although this may still leave the share of government investment in GDP below that in other major competitors. Probably the single area where decades of low government investment may be adversely impinging on the productivity performance of the whole economy is through an inadequate transport infrastructure. The latest comparative data show that UK roads are much more congested than those in other EU countries and that UK commuters are much more likely to encounter congestion on their way to work (CBI, 2003). As well as the cost of lost hours worked, by reducing journey time reliability, congestion risks further increasing business costs by under-mining just-in-time production techniques since around four-fifths of freight goes by road. The reliability of the railways also underperforms the other major European countries, as discussed in Chapter V. The government has announced a ten-year plan for major investments in roads and railways, but in the central projection road traffic is projected to rise by at least 20 per cent, with the likelihood that congestion will worsen rather than improve. Following the apparent success of the London congestion charge (Box 4.3), as much in terms of its public acceptance as its technological achievement, there is increasing interest in the possibility of the more widespread introduction of road charging. There is compelling evidence that distances travelled by

© OECD 2004


Box 4.3.

125

The London Congestion Charge*

London introduced a road charge of £5 a day for travelling in the central area in February 2003. There are no tollbooths or barriers around the charging zone, and no physical tickets or barriers. Instead drivers pay to register their vehicle registration number on a database for journeys within the zone. Cameras then capture images of vehicles entering, driving within or leaving the zone and their registration numbers are recognised and matched against those that have paid or are exempt by a computerised system. Images of vehicles where no charge has been paid are issued with a penalty charge notice. Initial results suggest traffic in the central zone during charging hours has fallen 16 per cent. Traffic congestion, measured in terms of minutes of delay per kilometre travelled compared to the travel rate in uncongested conditions, has fallen about 30 per cent. The standard deviation of average travel times, a measure of reliability, has also been reduced by 30 per cent. Trips around the zone boundary are generally flowing freely and additional buses have been able to carry passengers now choosing to use public transport. In addition, there has been a significant improvement in the reliability of buses across London as a consequence of reduced congestion in the centre. Net revenue raised in 2003-04 is expected to be £68 million, less than the £120 million initially expected, mainly because of a bigger than expected fall in traffic, although other factors such as a higher-than-expected number of exempt vehicles and evasion also contributed. Measures to tighten enforcement are expected to increase future revenues to around £100 million. * For further details see “Congestion Charging 6 Months On” Transport for London, October (2003), available from: www.tfl.gov.uk/tfl/.

road are sensitive to the costs of motoring (Glaister, 2002), and a recent study by the Independent Transport Commission suggested that there could be major gains, in terms of lower congestion as well as reduced pollution, from the widespread introduction of road charging (Glaister and Graham, 2003). Assessment While the UK has substantially closed GDP per capita gaps with the major European countries it should set its sights on closing the gaps with the more dynamic countries such as the United States, Canada and Australia which have been able to simultaneously raise labour utilisation and improve productivity growth, particularly through increased use of ICT. While overall levels of labour utilisation are high, this conceals a problem of growing numbers dependent on disability-related benefits which has barely

© OECD 2004

126


begun to be addressed as a policy issue. A new scheme, which has just begun to be piloted and involves mandatory job-focussed interviews for new applicants and a “return to work” credit, appears to be a promising approach. In any case the future direction of policy will require a substantial increase in the scale of welfareto-work programmes for the disabled, making it compulsory for the majority of new claimants with more manageable medical conditions. Consideration should also be given to applying any expanded scheme to the existing stock of claimants, not just new claimants. At the same time more resources should be devoted to vocational rehabilitation schemes and consideration given to whether temporary financial incentives for potential employers might be effective in initially getting those on incapacity benefits back into work. The other main New Deal schemes on the whole appear to be working well, with the rise in the employment rate for lone parents providing a striking example of the positive results that can be achieved by concerted policy action targeted at a particular disadvantaged group. However, the New Deal 50 plus should be monitored more closely, on a similar basis to other New Deal programmes, to ensure that it is cost effective, and a mandatory element introduced, as is being currently piloted. To further close the productivity gap with the better performing competitors, deficiencies in human and physical capital need to be addressed. Human capital is particularly important in explaining differences in growth performance among OECD countries. The United Kingdom compares poorly with its major competitors on both basic literacy and vocational skills suggesting this has been a major factor holding back productivity growth and may have prevented the full benefits from increased ICT investment from being reaped. The education system has improved recently, but many still leave education early and relatively few get an apprenticeship, skilled craft or technician level qualification. Training of persons with limited skills already in work is valuable from a personal as well as productivity perspective, but only if improvements are sufficient to warrant the costs of tuition and the loss of production from being out of work. There has been a pick-up in business investment per worker, which maybe partly attributable to greater macroeconomic stability, although it is still lower than in major competitor countries. There are sizeable investment gaps in residential and non-residential construction, which tends to confirm other evidence that strict planning controls are a problem. Government investment has also been unusually low in international comparison and, although a substantial increase is planned, it will still fall short of levels in most other countries. The need for increased public investment is most urgent in transport, but will not by itself solve congestion problems, but should be supplemented by the introduction of road charging schemes.

© OECD 2004


127

Notes 1. Of course, macroeconomic stability should itself contribute to potential output growth (OECD, 2003c). 2. In 2002 standardised inactivity rates for persons aged 15-64 was 23.4 per cent for the United Kingdom, compared to an EU average of 30.2, with the United States at 23.6, Canada at 22.5, and Australia at 26.1 per cent. Denmark, Norway, Sweden all had inactivity rates close to 20 per cent, whereas Iceland had the lowest inactivity rate at 14.4 per cent. 3. The fall in the female inactivity rate has coincided with a steady fall in the number of women reporting that they are “looking after family/home” (Jones et al., 2003). 4. The numbers claiming incapacity benefit are much higher than the number of recipients. This is because to receive incapacity benefit someone has to be both sufficiently ill or disabled and have paid sufficient national insurance contributions. Roughly onethird of claimants satisfy the first criterion but not the second, so do not receive incapacity benefit, but are then able to claim income support on the grounds of incapacity. 5. The Liverpool ward of Everton with a working age population of just over 3 000 recently had 320 unemployment benefit claimants and 1 130 people claiming disability related benefits (Financial Times, 2003). 6. A general finding of income-tested tax credits across a number of evaluations is that they may provide a disincentive for a second family member to earn additional income, particularly if they move the household into the withdrawal phase of the tax credit scheme where marginal effective tax rates are high (OECD, 2003f). 7. Indeed, there is evidence to suggest that a return to more normal routines including work is likely to be of benefit to such conditions. For references see the evidence summarised in Department for Work and Pensions (2002). 8. This difference is not explained by a larger group of immigrants with poor literacy in the United Kingdom. Among native-born adults alone, 21.5 per cent are at literacy level 1 in the United Kingdom, and again this is more than twice as many as in Germany, the Netherlands and the Scandinavian countries, and more than in Australia, Canada and the United States where 12.3, 14.8 and 17.5 per cent of the native-born adults are at literacy level 1. 9. Given the differences in national education systems, the categories used for international comparisons will always be a bit different from those used in each country. In the classification used by Crafts and O’Mahony (2001) and O’Mahony and De Boer (2002), for example, the United States has more low-skilled than the United Kingdom which in turn has considerably more low-skilled than France, because they classify persons with a US high school exam and with UK “A levels” as low-skilled. Irrespective

© OECD 2004

128

10.

11.

12.

13. 14.

15.


of which classification is used, however, the picture of the UK having relatively many low-skilled and few with intermediate skills is confirmed. As a manager expressed it in a survey: “There will always be room for the academic subjects, but we must have more of the basic, relevant skills that the economy needs” (The Daily Telegraph, 21 July 2003). The increasing supply of theoretically trained graduates lacking commercial understanding and work experience has led to a widening divergence of salaries and career prospects across the graduate labour market (Mason, 2001; 2002). A working group led by Mike Tomlinson is currently considering possible reforms of education for the 14-19-year olds. Its initial proposals, published in July 2003, include the creation of a diploma that will prepare young people for different types of further education, training and employment, thereby creating better pathways. For the student loan for living costs, introduced in 1998, repayments are already income contingent, as graduates pay 9 per cent of their income above a threshold of £10 000 a year, until the loan is fully amortised. The actual level will depend a lot on the extent to which universities raise their charges from the level of the current fees. In the Netherlands, individual loans fund part of tuition costs, and if the income of a graduate falls below a threshold she/he can request to pay less in the given year. This administrative procedure has to be renewed each year, although after 15 years any loan not repaid is forgiven. The Government has commissioned an independent review, being undertaken by Kate Barker, to consider the reasons for the lack of supply and responsiveness of housing in the UK. The Review’s recommendations are expected in spring 2004 (Box 2.1).

© OECD 2004


129

Bibliography Australian Ministry of Education (2003), HECS Information 2004, www.hecs.gov.au/hecs.htm (accessed on 9 December 2003). Barham, C. (2002a), “Economic Inactivity and the Labour Market”, Labour Market Trends, Office for National Statistics, February. Barham, C. (2002b), “Patterns of Economic Activity among Older Men”, Labour Market Trends, Office for National Statistics, June. Barr, N. (2001), The Welfare State as Piggy Bank, Oxford University Press. Barr, N. (2003a), Financing Higher Education in the UK: The 2003 White Paper, paper presented at the House of Commons Education and Skills Committee, Session 2002-03. http:// econ.lse.ac.uk/staff/nb/Barr_Selcom030311.pdf. Barr, N. (2003b), Financing Higher Education: Comparing the Options, 10 June 2003, London School of Economics and Political Science, http://econ.lse.ac.uk/staff/nb/barr_HE_option030610.pdf. Basu, S., J.G. Fernald, N. Oulton, S. Srinivasan (2003), “The Case of Missing Productivity Growth: Or Does Information Technology Explain why Productivity Growth Accelerated in the United States but not in the United Kingdom?”, NBER Macroeconomics Annual 2003, Volume 18. Blöndal, S., S. Field and N. Girouard (2002), “Investment in Human Capital through UpperSecondary and Tertiary Education”, OECD Economic Studies, No. 34. Blundell, R. and H. Reed (2000), “The Employment Effects of the Working Families Tax Credit”, Institute for Fiscal Studies Briefing Note, No.6, April. Blundell, R., M. Cost Dias, C. Meghir and J. Van Reenen (2002), “Evaluating the Employment Impact of a Mandatory Job Search Program”, Institute for Fiscal Studies, December. Blundell, R., H. Reed, J. Van Reenen, and Shephard (2003): “The Impact of the NDYP on the Labour Market; a Four Year Assessment”, In: Dickens, Gregg and Wadsworth (eds.), The Labour Market under New Labour, Palgrave, MacMillan. Brewer, M. (2003), “The New Tax Credits”, Institute for Fiscal Studies Briefing Note, No. 35, April. Card, D. and R. Freeman (2002), “What Have Two Decades of British Economic Reform Delivered”, National Bureau of Economic Research Working Paper, No. 8801, February. Carneiro, P. and J. Heckman (2003), “Human Capital Policy”, NBER Working Paper, No. 9495. Chapman, B. and C. Ryan (2002) “Income-Contingent Financing of Student Charges for Higher Education: Assessing the Australian Innovation”, Australian National University Discussion Paper, No. 229. Confederation of British Industry (2003), “Is Transport Holding the UK Back?”.

© OECD 2004

130


Crafts, N. and M. O’Mahony (2001), “A Perspective on UK Productivity Performance”, Fiscal Studies, Vol. 22, No. 3. Department for Education and Skills (2003a), 21st Century Skills, Realising Our Potential, July. Department for Education and Skills (2003b), The Future of Higher Education, January. Department for Education and Skills (2003c), Departmental Report 2003. Department for Work and Pensions (2002), “Pathways to Work: Helping People into Employment”, Cm 5690, HMSO, November. Department for Work and Pensions (2003a), “New Deal 50 Plus Quantitative Analysis of Job Retention”, Research Report, 151. Department for Work and Pensions (2003b), “New Deal for Lone Parents: Second Synthesis Report of the National Evaluation”, Research Report, 163. Dickens, R., P. Gregg, J. Wadsworth (2003), The Labour Market Under New Labour, Palgrave, Macmillan. Disney, R. and D. Hawkes (2003), “Declining Employment of Older Workers: Has Britain Turned the Corner?”, February. Glaister, S. (2002), “UK Transport Policy 1997-2001”, Oxford Review of Economic Policy, Volume 18, Issue 2, Summer. Glaister, S. and D. Graham, (2003), “Transport Pricing and Investment in England”, Report for the Independent Transport Commission, May, www.trg.soton.ac.uk/itc/reports.htm. Goodman, A. and G. Kaplan (2003), “Study Now, Pay Later” or “HE for Free?”An Assessment of Alternative Proposals for Higher Education Finance, Institute for Fiscal Studies Commentary, No. 94. Goodman, A. and L. McGranahan (2003), “Access to Education”, in Dickens, R., P. Gregg and J. Wadsworth (eds.) The Labour Market under New Labour, Palgrave Macmillan. Gregg, P. and J. Wadsworth (1999), “Economic Inactivity”, in P. Gregg and J. Wadsworth (eds.), The State of Working Britain, Manchester, Manchester University Press. Gregg, P. and J. Wadsworth (2003), “Workers Households and the Recovery” in Dickens, R., P. Gregg and J. Wadsworth (eds.) The Labour Market under New Labour, Palgrave Macmillan. Gregg, P. and J. Wadsworth, (2000), “Mind the Gap, Please: The Changing Nature of Entry Jobs in the UK”, Economica, 67. Gregg, P. and S. Harkness (2003), “Welfare Reform and the Employment of Lone Parents in the UK”, in The Labour Market Under New Labour, R. Dickens, P. Gregg and J. Wadsworth (eds.), Palgrave Macmillan. Gust, C. and J. Marquez (2002), “International Comparisons of Productivity Growth: The Role of Information Technology and Regulatory Practices”, Labour Economics, Special Issue on Productivity edited by Gilles Saint-Paul. H.M. Treasury and Department for Work and Pensions (2001), “The Changing Welfare State: Employment Opportunities for All”, available from, www.hm-treasury.gov.uk, November. H.M. Treasury (2003a), “Pre-Budget Report”, www.hm-treasury.gov.uk/. H.M. Treasury (2003b), “Lambert Review of Business-University Collaboration, Final Report”, December 2003, www.hm-treasury.gov.uk/consultations_and_legislation/lambert/. IFS (2003), IFS Analysis of Higher Education Funding Options: Fact Sheet, www.ifs.org.uk/press/ tuition_fees.pdf (accessed on 8 December 2003). International Monetary Fund (IMF) (2003), United Kingdom, Selected Issues, February.

© OECD 2004


131

Jones, J., M. Joyce and J. Thomas (2003), “Non-Employment and Labour Availability”, Bank of England Quarterly Bulletin, Autumn. Kaplan, G. and A. Leicester (2002), “A Survey of the UK Benefit System”, Institute for Fiscal Studies Briefing Note, No. 13, October. Martin, J. and D. Grubb (2001), “What Works and for Whom; A Review of OECD Countries’ Experiences with Active Labour Market Policies”, IFAU Working Paper, 2001:14. Mason, G. (2001), “The Mix of Graduate and Intermediate-level Skills in Britain: What Should the Balance Be?”, Journal of Education and Work, Vol. 14, No. 1. Mason, G. (2002), “High Skills Utilization Under Mass Higher Education: Graduate Employment in Service Industries in Britain”, Journal of Education and Work, Vol. 15, No. 4. Mason, G. and K. Wagner (2002), Skills, Performance and New Technologies in the British and German Automotive Components Industries, Report prepared for the Department for Education and Skills. New Zealand Ministry of Education (2003), Student Support in New Zealand, September 2003. O’Mahony, M. and W. De Boer (2002), “Britain’s Relative Productivity Performance: Has Anything Changed?”, National Institute Economic Review, No. 179, January. OECD (2002a), “Society at a Glance: OECD Social Indicators”, OECD, Paris. OECD (2002b), Mechanisms for the Co-finance of Lifelong Learning, www.oecd.org/dataoecd/51/ 27/2501342.pdf. OECD (2002c), Economic Survey of New Zealand, Paris. OECD (2003a), “Transforming Disability into Ability: Policies to Promote Work and Income Security for Disabled People”, OECD, Paris. OECD (2003b), “Making Work Pay – Making Work Possible”, Chapter 3, OECD Employment Outlook, OECD, Paris. OECD (2003c), The Sources of Economic Growth in OECD Countries, OECD, Paris. OECD (2003d), “Upgrading Workers’ Skills and Competencies”, Chapter 5 in OECD Employment Outlook, OECD, Paris. OECD (2003e), Beyond Rhetoric: Adult Learning Policies and Practices, OECD, Paris. OECD (2003f), “The Labour Mobilisation Challenge: Combating Inactivity Traps and Barriers to Moving up Job Ladders”, Chapter 2, OECD Employment Outlook, OECD, Paris. OECD (2003g), “Taxing Wages 2001-2002”, OECD, Paris. OECD and Statistics Canada (2000), Literacy in the Information Age, Final Report of the International Adult Literacy Survey. Ok, W. and P. Tergeist, (2003) “Improving Workers’ Skills: Analytical Evidence and the Role of the Social Partners”, OECD Social, Employment and Migration Working Papers, No. 10. www.oecd.org/dataoecd/60/6/5401597.pdf. Stenberg, A. (2003), “Comprehensive Education for the Unemployed – Evaluating the Effects on Unemployment of the Adult Education Initiative in Sweden (Kunskapslyftet)”, Umeå Economic Studies, No. 579. Sweden’s Centrala studiestödsnämnden (2003), Studiemedelsguiden, March 2003 www.cse/BlanketterOchInformationsmaterial/Blanketter.asp?MenyIdnr=223 accessed on 9 December 2003). UCAS (Universities and Colleges Admissions Service) (2000), Statistical Bulletin on Widening Participation, Cheltenham.

© OECD 2004

132


UCAS (2002), Paving the Way, project report, May 2002, www.ucas.com/candq/paving/index.html (accessed 10 December 2003). Van Reenen, J. (2001), “No More Skivvy Schemes? Active Labour Market Policies and the British New Deal for the Young Unemployed in Context”, Institute for Fiscal Studies Working Paper, 01/09, May. Westergaard-Nielsen, N. (2001), “Danish Labour Market Policy: Is it Worth it?”, Centre for Labour Market and Social Research (CLS), Working Paper, No. 01-10, November 2001. York Consulting (2002), Individual Learning Accounts – Follow Up Study, DfES Research Brief, No. RBX 01-02, January, www.dfes.gov.uk/research/data/uploadfiles/RBX01-02.doc.

© OECD 2004


133

Annex 4.A1.

Funding of tertiary education in OECD countries While participation in tertiary education in nearly all OECD countries rose during the second half of the 1990s, expenditure on tertiary education as a share of GDP was, on average, little changed.1 This partly reflects funding mechanisms that were designed for a different era and which constrain the amount of funding coming from non-government sources, while at the same time competing demands on aggregate public expenditure may have squeezed out public finance of tertiary education. This annex reviews tuition fees and public loan schemes for tertiary education in OECD countries, in order to put the recent UK proposals for the graduate contribution scheme in international context. Variability of tuition fees Any increase in the total expenditure going into tertiary education is likely to involve an increase in tuition fees. The funding for UK tertiary education institutions coming from tuition fees paid by students is equal to 0.15 per cent of GDP, much smaller than in Korea, the United States, Japan, New Zealand, Canada and Australia, although quite high relative to many European countries, where tuition fees only exist for limited parts of tertiary education (Figure 4.A1.1). Tuition fees can vary across subjects and across universities for the same subject or degree.The United Kingdom has so far applied a common level of fees across subjects and universities,2 whereas most of the other OECD countries charging tuition fees (with some exceptions such as the Netherlands) allow some variation. The UK government’s proposals would allow universities to charge up to £3 000 per annum per student,3 which could increase total tertiary education funding by up to 0.15 per cent of GDP, raising expenditure on tertiary education expenditure by around one-seventh relative to current levels.4 In countries where the level of tuition fees varies across subjects, it is typically low in arts and high in medicine with science and law courses varying between these two extremes (for example in Australia, Ireland and New Zealand). This is arguably both fair and efficient, given the higher costs of such courses and that graduates from these courses typically earn higher incomes. Nevertheless, there are difficult questions concerning whether there is a case for subsidising particular courses which might be judged as important for promoting growth. For example, the New Zealand government is considering re-orienting public funding in order to address the relative shortfall of engineering, mathematics and computer-science students compared with other OECD countries (OECD, 2002c). But even in such cases it would appear preferable if the extent of any subsidy is explicit, rather than implicit because a uniform fee is imposed across all courses. In those countries where fees can vary across universities for the same subject or degree, very different outcomes can be observed. For example, in the United States fees vary substantially across universities, as notably private universities charge fees that are several

© OECD 2004


134

Figure 4.A1.1.

Tuition fees in tertiary education 2000

Per cent of GDP

Per cent of GDP

1.6

1.6 Tuition fees Tuition fees paid by government

1.4

1.4

1.2

1.2

1.0

1.0

0.8

0.8

0.6

0.6

0.4

0.4

0.2

0.2

0.0

JAP 1

KOR USA 1

CAN 1 NZD

IRE 1 AUS

UK 2 ESP

ITA MEX

ISL 1

FRA NLD

PRT 1

NOR DNK

0.0

GRC AUT

SWE

2. No information on how much of tuition fees are paid by government. 3. 1999. Source: OECD, Education Database and Statement of Financial Performance for tertiary education institutions in New Zealand (see footnote 1 of Figure 4.7).

times higher than the average of about USD 6 000. The level of tuition fees in the US public sector is typically decided by the university itself, but in many states there are regulations limiting the level of fees charged for students coming from within the state. Over the 1990s, the level of tuition fees rose rapidly to offset declining public funding, illustrating that variation in tuition fees can provide an important safety valve during periods of fiscal consolidation. Availability of student loans Many OECD countries operate some form of public student loans to finance tuition fees and/or living costs. However for only few countries including Australia, Iceland, New Zealand, Norway and Sweden do student loans amount to 0.2 per cent of GDP or more, with New Zealand having the highest level of student loans equivalent to more than ½ per cent of GDP (Figure 4.A1.2). Student loans are in some cases tied to the payment of tuition fees, as with the Higher Education Contribution System in Australia or the proposed Graduate Contribution Scheme in the United Kingdom. Alternatively, as in the Netherlands, New Zealand and the United States, loans schemes help students finance the tuition fees, but lending is not tied to tuition fees, and the same conditions apply whether a student borrows to finance tuition fees or living costs (Table 4.A1.1). In some countries the limited use of loans should be seen in the context of significant public grants, in some cases targeted at students from low-income families. Moreover, the dividing line between loans and grants can be blurred, as in the Netherlands, where all students living away from their parents receive a basic grant of about EUR 2 600, and a supplementary grant is available for students from low-income families. Both grants are initially

© OECD 2004


135

Figure 4.A1.2. Public loans to students in tertiary education 2000 Per cent of GDP

Per cent of GDP

0.6

0.6

0.5

0.5

0.4

0.4

0.3

0.3

0.2

0.2

0.1

0.1

0.0 NZL SWE NOR

ISL

AUS

UK1 DNK CAN

0.0 USA NLD

JPN

DEU KOR MEX CHE

ITA

FIN

IRE

ESP

1. 1999. Source: OECD, Education Database and Statement of Financial Performance for tertiary education institutions in New Zealand (see footnote 1 of Figure 4.7).

made as loans that are then converted to grants providing students meet minimum academic criteria, which most do. Insurance via income contingency for repayments A way of reducing the risk carried by individuals investing in their own education is to make student loan repayments conditional on graduate income. In Australia, New Zealand as well as in the United Kingdom with both the current maintenance loan and the proposed Graduate Contribution Scheme, the speed of repayments depends on graduate income (Table 4.A1.1). Hardship procedures may to some extent replicate income contingency, and exist in most countries, such as the Netherlands, where persons with low income can request temporarily reduced repayments.5 Income contingent repayments, however, can be thought of as a less bureaucratic alternative to hardship procedures. Moreover, in the United States where the obligation to repay loans does not depend on graduate income, the default rate is high, in particular for persons having studied at vocational institutions. In Sweden, repayments on loans to cover living costs used to be 4 per cent of a graduate’s income. Now, although repayments will be initially independent of income, graduates are entitled to have repayments reduced to equal no more than 5 per cent of their income. This, however, does not reduce the period over which loans are to be repaid, and consequently the borrower will have to pay more in later years. The income threshold at which repayment begins for the UK maintenance loans is currently £10 000, which at roughly half the gross earnings of the average production worker,6 is close to the thresholds applied in Australia for the Higher Education Contribution System and in New Zealand (Table 4.A1.1). As part of the proposed changes this threshold will rise

© OECD 2004


136

Table 4.A1.1. Funding of tertiary education in OECD countries Countries having fees and loans with income-contingent repayment

Total public and private expenditure for educational institutions, % of GDP Public Direct expenditure for institutions Grants to help students pay tuition fees

Australia

Netherlands

New Zealand

United Kingdom1

Sweden2

United States2

1.55

1.23

1.98

1.03

1.61

2.66

0.80

0.98

0.98

0.76

1.42

0.90

0.79

0.95

0.98

0.65

1.42

0.90

0.01

0.03

0.00

0.11

0.00

..

0.75

0.25

1.00

0.27

0.19

1.76

0.48

0.11

0.55

0.15

0.00

0.79

Private Tuition fees paid by students (net of grants above) Other private sources including donations

0.26

0.14

0.45

0.12

0.19

0.97

Public student loans3

0.20

0.09

0.54

0.14

0.40

0.09

Public grants to students for living costs

0.15

0.24

0.25

0.13

0.19

0.00

Tuition fees Can fees vary across ... ... subjects ... universities

Yes No4

No No

Yes Yes

No5 No5

.. ..

Yes Yes

Financing/loan schemes ... tied to tuition fees

None None Proposed:5 No fees Higher Graduate Education contributions Contribution System6 Income contingency Progressive Repayment of repayments rates on of 9% total of income income, over 75% if income of APW above 55% of APW Interest rate Zero real Zero real Maintenance ... untied, including living Financial loans costs Supplement loans7 Income contingency No Not Repayment Repayment Can reduce repayments of repayments repayments automatic, by 10% by 9% while but can of income of income temporarily to 5% of income is request over 40% over 50% income9 below 75% respite if low of APW of APW8 of APW income

None

Federal10 scheme No

© OECD 2004


137

Table 4.A1.1. Funding of tertiary education in OECD countries (cont.) Countries having fees and loans with income-contingent repayment Australia

Interest rate

Netherlands

Zero real Approximate market rate

New Zealand

United Kingdom1

Zero nominal while student, thereafter 7%

Zero real

Sweden2

United States2

Slightly Slightly subsidised subsidised

Note: Expenditure data refer to 2000; Qualitative information refer to academic year 2003/04 or latest available year. 1. Expenditure data refer to 1999 for the whole of the United Kingdom, as the data for 2000 on student loans are incomplete. 2. The United States and Sweden are included as a basis for comparison because of their particular experience although the United States does not have income-contingent loans and the Swedish loans fund only living costs. 3. Reflects gross lending to students. It is not possible to distinguish loans tied to payment of tuition fees from loans to finance living costs. 4. The Australian government has recently announced partial deregulation of tuition fees. 5. As part of the proposed changes, fees will be allowed to vary across subjects and universities up to a ceiling of £3 000. The details of the future Graduate Contribution Scheme which is to become operable from 2006 may still change as the proposal goes through Parliament. 6. Similar conditions apply for the Postgraduate Education Loan Scheme and the Bridging for Overseas-Trained Professionals Loan Scheme. 7. Financial Supplement Loans will not be available after 2003. 8. The threshold for repayment will increase to 75% of APW gross wage earnings from 2005. 9. Refer to loans issues since June 2001. 10. Many states have schemes that supplement and may differ from the federal scheme described here. Data in the table include both federal and state loans. Source: OECD, Education Database, and Statement of Financial Performance for tertiary education institutions in New Zealand. Descriptions of tuition fees and loan schemes is based on official publications from member countries including the international comparisons in New Zealand Ministry of Education (2003) and on Barr (2001).

to £15 000 and be equal to the threshold that will apply for the Graduate Contribution Scheme, making the insurance provided by income contingency greater than in other OECD countries. Interest rate subsidies Some student loan schemes imply a significant subsidy from low interest rates. In New Zealand, an interest rate similar to the market rate was applied to student loans until 1999, whereas today loans are interest free while the person is still a student. This implies a subsidy of a magnitude similar to one-third of the 0.54 per cent of GDP lent to students, illustrating how expensive it can be to have below-market interest rates. Applying a zero real interest rate throughout the loan period as currently with the maintenance loans in the United Kingdom and suggested for the Graduate Contribution Scheme implies an even larger subsidy, equal to about half the value of lending.7 Broad interest rate subsidies, however, do not necessarily help those who need it most, and applying an approximate market rate as in the Netherlands (2 percentage points above long-term government bonds), or a slightly subsidized rate as in the United States (which has variable rates capped at 8¼ per cent) and Sweden (Table 4.A1.1) would therefore free substantial resources for pro-access policies that could have a much bigger impact on improving equity of access.

© OECD 2004

138


Effects on access equity A central concern for the design of education funding is how it affects equity of access. As argued in Chapter 4, the main determinant of who will benefit from post-compulsory education is likely to be cognitive developments in early childhood and the foundation laid during compulsory education. But even focusing narrowly on the effects of letting students or graduates contribute to the cost of tertiary education, most studies find little or no sign of adverse equity effects from well designed schemes. New Zealand is a particularly clear example. Since the introduction of student loans in 1992, participation in tertiary education has almost doubled, and the number of young people entering tertiary education is today the highest in any OECD country. Moreover, the share of students from Maori and Pacific ethnic groups increased from 9 per cent in 1990 to 24 per cent in 2001. A key reason for this expansion has been that loans were available for any approved tertiary institution and that public funding for tertiary education institutions was reformed towards an equal per-student funding for public and private institutions. That led to an explosion of small, innovative vocational education providers who better reached students from non-traditional ethnic and socio-economic backgrounds (OECD, 2002c). On the other hand, countries trying to widen access by introducing or increasing grants to cover student living costs have had only little success. When back in the mid-1970s Australia abolished tuition fees and introduced grants, it had only little effect on the socioeconomic mix of students. Nor has the socio-economic mix of students changed following the re-introduction of tuition fees in 1989 based on income-contingent loans (Blöndal et al., 2002; Chapman and Ryan, 2002). High up-front tuition fees, however, requires students to finance tuition by private borrowing to the extent that they cannot cover the fees via support from their family or by working while studying. In the United States, the 0.09 per cent of GDP made available via public student loans is supplemented by an estimated 0.23 per cent of GDP via loans from private entities (some of which have a government guarantee), but the sum of these lending streams is still significantly lower than the tuition fees paid by students amounting to 0.79 per cent of GDP, which could indicate that students are credit constrained. In other countries the extent of lending to students from private entities is smaller. Studies from the UK Universities and Colleges Admissions Service indicate that while the tuition and living costs of taking tertiary education (and the associated accumulation of debt) is a source of concern for potential applicants (UCAS, 2002), the replacement of grants with loans and the introduction of tuition fees in 1998 did not affect the social-class mix of entrants to university adversely, and the proportion of ethnic minority entrants and women has grown since then (UCAS, 2000).8 Overall, the experience of countries that have combined an increase in tuition fees with an increase in student loan facilities suggests that there are no significant adverse effects on participation, as concluded by Blöndal et al. (2002).

© OECD 2004


139

Notes 1. For figures on the rise in participation in tertiary education see indicator C2 of OECD Education at a Glance. For figures on the share of expenditure on tertiary education in GDP see indicator B2 of OECD Education at a Glance. For the 25 OECD countries for which data is reported for both 1995 and 2000, the (unweighted) average of expenditure on tertiary education as a percentage of GDP is the same in both years. In the United Kingdom, enrolment in tertiary education has almost doubled since 1989, but this has not been matched by spending increases, and consequently spending per student has declined by 37 per cent in real terms between 1989 and 2002. 2. Means testing, however, makes about half of all students exempt from the current fee, in which case universities receive a fee payment from government as illustrated in Figure 4.A1.1. 3. For the first £1 125 the current means testing would continue to apply. 4. This estimate is based on information in a government written answer to a parliamentary question (Hansard 12 November 2003, Column 354W), indicating that if all universities go to the maximum fee level of £3 000 then the extra revenue will amount to £1.4 billion for England which is about 0.15 per cent of England’s GDP. This is in addition to about £0.5 billion currently paid in up-front fees that in the future will also be paid via the Graduate Contribution Scheme. Together, this equals 0.2 per cent of England’s GDP. Growing student numbers may raise this figure in the future. 5. In New Zealand, persons with high debt relative to their income get part of the interest written off which adds to the insurance from income-contingent repayments. 6. Data on the gross earnings of the average production worker for OECD countries in 2002 can be found in OECD (2003g). 7. The extent of subsidy will depend on how real interest rates evolve on capital markets in the future and can therefore only be roughly assessed now (IFS, 2003). 8. As the fees introduced in 1998 apply only to students from families with income over a certain threshold, their introduction could in principle have led to a wider social-class mix of student if persons having to pay the fee had felt discouraged. In practice, this effect has been insignificant.

© OECD 2004

V.

Product market competition and economic performance

Overview There is a well-identified empirical connection between the intensity of competition in product markets and better productivity performance (OECD, 2002a). In general, competitive pressures appear to be relatively strong in the United Kingdom but this does not mean that there is no room for improvement. The overhaul in competition legislation that occurred in recent years was long overdue and active enforcement along with the new market studies and market reference regimes of the competition enforcement agencies should help to ensure that markets are competitive. In service sectors the United Kingdom has experienced relatively strong labour productivity growth over the 1990s and a large part of this is due to the introduction of increased competition over this period. However, market power on the part of incumbent firms in some sectors remains a concern and current planning restrictions are inhibiting competition in the retail sector. While recent steps by the government overcome the most serious weaknesses of the privatised rail system, problems regarding incentives and clear lines of responsibility remain to be resolved. This chapter assesses what role product market competition, and the policies that impact upon competition, may have played in the performance of the UK economy over the past decade and what further measures might contribute to enhancing growth. The main links between stronger competition and macroeconomic performance are reviewed in the first section of this chapter, while the second section lays out the competition legislation framework and the recent introduction of the Enterprise Act. In the third section, competitive conditions, regulation and recent reforms are analysed for a wide range of service sectors. This includes service industries that are competitive, such as retail distribution and professional services, and network industries containing non-competitive segments, such as telecommunications, electricity and rail. A concluding section draws on the analysis and provides a set of policy recommendations that highlight areas in which stronger competition can boost performance.

© OECD 2004

142


Product market competition and macroeconomic performance The United Kingdom’s average annual productivity growth (per worker) between 1990 and 2001 is slightly higher than the OECD and EU averages, and the highest amongst the G7 countries (Table 5.1). However, superior annual productivity growth of the order of a few tenths of a percentage point needs to be put into context: the magnitude of the productivity gap with the United States and the major continental European countries is still around 20 percentage points or more, although the gap with the latter has been closing (Chapter I). A sectoral comparison of labour productivity growth shows that the United Kingdom had the highest productivity growth of the G7 countries in construction and phenomenal rates in electricity, gas and water, where annual productivity growth averaged just over 10 per cent a year, primarily due to liberalisation and regulatory reforms undertaken in electricity and gas over the past decade. In relation to the G7, productivity performance in transport and communications was relatively good, but average in manufacturing. Although productivity growth in distribution was higher than in most other major countries, perhaps more tellingly it was only half the rate of growth achieved in the United States. Overall, while productivity growth over the last decade appears encouraging, there is considerable scope for reducing the productivity gap with major competitors further. Indicators of the intensity of product market competition Overly stringent product market regulations can have an impact on the strength of competition in domestic markets either by exerting direct control on economic activities or by maintaining high barriers to trade and foreign direct investment. Various barriers to entrepreneurial activity (e.g. restrictions on market access or administrative burdens and red-tape on firms) can also inhibit competition and discourage entry by both domestic and foreign firms. In international comparison the United Kingdom’s institutional and regulatory arrangements are in general among the most favourable to competition. For example, a recent study by the European Commission (EC) shows that, both in terms of cost and time, British firms incurred the lowest costs and faced the lowest waiting times for starting up a new business (EC, 2002a). And focusing on regulations that restrict competition and market mechanisms (e.g. economic and administrative regulations and barriers to trade and foreign direct investment (FDI), OECD indicators of regulation show that in 1998 the United Kingdom had the least restrictive economy-wide regulatory stance in the OECD area (Figure 5.1).1 Contributing to the low economywide indicator are the reforms that the United Kingdom undertook in the 1980s and 1990s in the utilities and transport sectors.2 Although it is difficult to classify markets according to the strength of market forces, the degree of product market competition may be gauged from jointly considering a number of imperfect proxy measures. The measures of product

© OECD 2004

Output, employment and productivity 1990 to 2001

Average GDP growth of which: Labour productivity Employment Labour productivity growth in selected industries Total manufacturing Electricity, gas and water supply Construction Wholesale and retail trade; restaurants and hotels Transport and storage and communication

United States

Canada

Australia

2.8

2.5

3.3

1.7

1.6 1.2

1.3 1.2

1.9 1.4

3.3 1.2 0.0 3.3

2.2 0.9 0.3 1.3

2.7

3.0

1. Average growth 1992-2001. 2. OECD except Austria, Luxembourg, Poland and Turkey. 3. EU except Austria and Luxembourg. Source: OECD.

Japan

Germany1

United Kingdom

OECD2

European Union3

2.3

2.5

2.0

1.6 0.1

1.8 0.4

1.6 0.9

1.5 0.6

3.3 3.0 0.1 0.8

1.9 3.4 0.0 1.0

2.6 10.1 2.5 1.7

n.a n.a n.a n.a

n.a n.a n.a n.a

2.6

3.6

3.9

n.a

n.a

France

Italy

1.5

1.9

1.7

1.3 0.4

1.7 –0.2

1.1 0.8

2.4 5.9 0.6 1.9

2.7 2.6 –2.4 2.2

2.5 5.1 0.0 –0.6

4.4

1.5

7.3


© OECD 2004

Table 5.1.

143


144

Figure 5.1. Indicators of product market regulation1

2.5

A. Overall regulatory approaches by area, 1998 2

2.5 Economic regulation 3 Administrative regulation Barriers to trade4

2.0

2.0

1.5

1.5

1.0

1.0

0.5

0.5

0.0 8

ITA

GRC

NOR

FRA

BEL

CHE

PRT

FIN

ESP

JPN

CAN

DEU

DNK

AUT

NLD

SWE

NZL

USA

AUS

0.0 UK

8

B. Change in regulatory stance 5 1978

6

6

4

4

2

2 1998

0

0 CHE

GRC

ITA

ESP

CAN

PRT

AUT

FRA

NLD

SWE

JPN

BEL

NOR

USA

DNK

DEU

AUS

FIN

UK

NZL

1. The regulatory stance is measured by a synthetic indicator ranging between 0 (least restrictive) and 6 (most restrictive). 2. Indicator of economy-wide product market regulations. 3. Includes barriers to competition and state control. 4. Includes trade and FDI restrictions. 5. Reports changes in the regulatory stance in seven non-manufacturing industries (gas, electricity, post, telecommunications, passenger air transport, railways and road freight) between 1978 and 1998. Source: Nicoletti et al. (2001); Nicoletti and Scarpetta (2003).

market competition presented below primarily look at the manufacturing sector but some non-manufacturing sectors are also considered. Manufacturing industries are grouped into four categories. A distinction is made between low R&D and high R&D industries; and between fragmented industries, which are those industries that are less concentrated and characterised by a large number of firms, and segmented industries, which are more concentrated industries characterised by relatively large firms. Structural measures such as industry concentration ratios or indices are often used as an indicator of competitive forces. Hirshman-Herfindahl indices show that, in general, UK industry (both manufacturing and services) is somewhat less concentrated than in comparable countries like France and Italy (Table 5.2). In general, mark-ups, a frequently used gauge of market power, and thus competitive pressures, appear to be relatively low in UK manufacturing compared with other

© OECD 2004


145

Table 5.2. Hirshman-Herfindahl indices of industry concentration United Kingdom 2000

France 1999

Manufacturing industry segmented, high R&D Chemicals products Office and computing machinery Electrical machinery Radio, TV and communication equipment Motor vehicles Other transport equipment

43.5 284.8 31.0 133.6 89.7 116.6

30.0 1 292.5 97.2 226.5 139.9 339.2

38.6 710.6 23.3 106.8 148.2 289.8

Fragmented, high R&D Medical appliances, optical instruments, watches and clocks Machinery and equipment Furnitureand other manufacturing

42.9 16.9 13.1

48.5 22.1 52.1

24.0 8.9 5.6

Segmented, low R&D Coke and petroleum products Basic metals Plastic and rubber products Food and beverages Tobacco products

n.a. 111.6 21.7 27.2 n.a.

753.1 114.7 48.1 14.9 n.a.

830.4 60.1 24.5 11.7 2 334.5

Fragmented, low R&D Textiles Wearing apparel Leather and footwear Wood products Paper and pulp products Publishing and printing Non-metallic products Fabricated metal products

19.4 29.0 134.0 15.6 78.8 14.1 44.4 8.6

20.5 23.8 58.0 14.3 63.2 15.4 59.6 9.0

8.2 18.1 11.2 3.4 76.9 46.6 25.6 1.6

Non-manufacturing industry Electricity and gas Water Land transport Water transport Air transport Post and telecommunications Construction Sale and repair of motor vehicles Wholesale trade Retail trade Hotels and restaurants

n.a. n.a. 16.4 101.5 144.3 105.5 4.0 17.2 3.4 13.6 11.3

1 327.9 709.5 182.0 471.3 1 375.7 1 186.8 2.3 7.0 2.6 10.5 5.8

436.7 151.5 56.8 112.0 n.a. 825.2 0.7 n.a. n.a. n.a. 4.1

Source:

Italy 1999

OECD, Statistics on enterprises by size class (SEC database).

countries (Figure 5.2). According to this indicator, competition pressures in manufacturing seem to be quite intense in all market structure classifications. Lower than average estimated mark-ups imply that manufacturing firms in the United Kingdom have relatively littlemarket power.

© OECD 2004


146

Figure 5.2. Average mark-ups by market structure 1981 to the latest available year1

1.5

1.5

A. Fragmented, low R&D

1.4

1.4

Average

1.3

1.3

1.2

1.2

1.1

1.1

1.0

1.5

BEL

CAN

FIN

FRA

DEU

ITA

JPN

NET

UK

USA

1.5

B. Fragmented, high R&D

1.4

1.0

1.4

Average

1.3

1.3

1.2

1.2

1.1

1.1

1.0

1.5

BEL

CAN

FIN

FRA

DEU

ITA

JPN

NET

UK

USA

1.0

1.5

C. Segmented, low R&D

1.4

Average

1.4

1.3

1.3

1.2

1.2

1.1

1.1

1.0

1.5

BEL

CAN

FIN

FRA

DEU

ITA

JPN

NET

UK

USA

1.0

1.5

D. Segmented, high R&D

1.4

1.4

Average

1.3

1.3

1.2

1.2

1.1

1.1

1.0

BEL

CAN

FIN

FRA

DEU

ITA

JPN

NET

UK

USA

1.0

1. The average mark-up is an un-weighted average of the available mark-ups, based on Roeger’s method (Oliveira Martins et al., 1996). ISIC, Rev. 3 classification. 1981-2000: Belgium and Finland, 1981-99: France, Netherlands and United States, 1981-98: Italy, Japan and United Kingdom, 1981-96: Canada, 1992-99: Germany. Source: OECD STAN database.

© OECD 2004


Table 5.3.

147

Import penetration by manufacturing industry Latest available year

Total manufacturing

Canada

France

Germany

Italy

Japan

United Kingdom

United States

1999

2000

2000

2000

2001

2000

2001

55.4

36.61

40.81

30.6

11.6

44.8

23.1

Fragmented, low R&D Textiles Wearing apparel Leather products and footwear Wood products Paper and pulp Printing and publishing Non-metallic products Fabricated metal products

65.0 42.7

52.11 53.4

88.1 77.1

20.0 23.2

35.6 33.8

54.0 65.2

27.2 53.3

81.8 19.3 .. .. 40.9 35.8

80.6 22.9 38.9 9.01 18.91 14.3

89.2 20.7 42.7 7.1 18.4 15.0

37.9 17.8 27.4 7.0 8.7 7.3

57.2 25.6 4.6 1.8 4.8 3.9

86.2 32.0 35.2 8.0 17.9 16.7

79.5 12.7 10.4 2.5 13.5 8.9

Fragmented, high R&D Medical precision and optical instruments Machinery and equipment Other manufacturing

.. 84.1 56.8

44.7 54.0 33.4

58.9 36.2 38.8

59.9 37.4 19.5

68.1 8.3 9.9

63.5 57.0 38.8

22.7 26.4 38.5

Segmented, low R&D Refined petroleum, coke Basic metals Shipbuilding and repairs Rubber and plastic4 Food, beverages, tobacco4

13.2 41.9 52.9 46.3 18.6

18.11 45.7 11.8 30.9 18.51

33.2 47.5 32.42 29.2 19.0

17.1 42.7 18.92 22.0 19.0

10.8 6.7 2.93 4.7 11.2

21.4 46.9 21.0 26.6 20.5

17.9 22.0 6.8 12.1 6.3

60.3 56.6

49.9 42.6

53.2 60.32

44.5 33.12

12.1 8.83

53.3 63.1

20.4 18.7

113.9 91.0

110.6 47.5

107.3 33.1

97.82 20.72

28.83 12.03

105.7 53.6

68.2 46.6

97.5 34.6 120.22 39.72 98.3

61.0 58.5 69.52 29.82 54.6

14.73 3.73 43.63 186.93 32.63

90.8 59.0 87.4 43.6 72.2

41.8 36.5 30.2 20.3 25.4

Segmented, high R&D Chemicals Pharmaceuticals Office and computing machines Electrical machinery Radio, TV and communication equipment Motor vehicles Aircraft Railroad equipment Other transport equipment

67.4 77.3 87.3 36.6 70.2

73.6 36.51 49.4 51.7 43.31

Note: Imports as a percentage of domestic demand (estimated as production minus exports plus imports). Values greater than 100 can occur when exports exceed production because of the inclusion of re-exports, i.e. products that are imported and then re-exported without any further transformation. “. .” indicates not available. 1. 2001. 2. 1999. 3. 2000. 4. Mix of fragmented and segmented sectors. Source: OECD STAN database.

© OECD 2004


148

The strength of competitive pressure also depends to a large extent on how exposed industries are to international competition. Import penetration ratios indicate that British producers face relatively stronger competitive pressure from foreign firms than their counterparts in other large European countries (Table 5.3).3 Among comparably sized economies, only Canada has a higher import penetration rate in total manufacturing. A sectoral breakdown shows that competitive pressures are across the board and reflect the fact that the United Kingdom is a relatively open economy. FDI inflows are also an indicator of the relative openness of an economy. The United Kingdom has comparatively high inflows and outflows of FDI, with both higher inflows and outflows as a percentage of GDP of any of the G7 countries (Figure 5.3). The relatively high import penetration Figure 5.3. Foreign direct investment outflows and inflows Per cent of GDP, average 1997-2001

20

(37.1)

20

A. Outflows

15

15

10

10

5

5

0

0

BEL-LUX FIN CHE FRA ESP CAN DEU ISL USA JPN HUN NZL TUR CZE NLD SWE UK DNK IRE PRT NOR AUT ITA AUS KOR GRC SVK POL

20

(38.1)

20

B. Inflows

15

15

10

10

5

5

0

0

BEL-LUX SWE SVK DNK FIN CHE HUN NZL ESP AUT FRA AUS KOR TUR JPN IRE NLD CZE UK CAN POL PRT DEU NOR MEX USA ISL GRC ITA

Source: OECD, International Direct Investment Statistics.

© OECD 2004


149

rates and FDI flows, in conjunction with lower than average mark-ups suggests a lack of barriers to entry and the existence of competitive pressures on manufacturing firms. Competition, innovation and economic performance Competition is generally considered a primary driver of innovative activity. A more competitive environment tends to strengthen R&D and diffusion of technologies, both of which are primary factors contributing to economic growth (Ahn, 2002; OECD, 2003a). Yet despite its relatively open and competitive environment, the United Kingdom is one of the few countries where R&D expenditure as a percentage of GDP has consistently fallen over the past two decades, falling by 25 per cent over the 1981 to 2000 period (Table 5.4). While R&D expenditure in the United Kingdom at the beginning of the 1980s was well above both OECD and EU averages, by the end of the 1990s it was just below the EU average and had fallen well below the OECD average. The fall in R&D spending is partly explained by the falling share of manufacturing in the UK economy (Chapter I).

Table 5.4. Gross domestic expenditure on R&D as a percentage of GDP 1981-2001 1981

1991

1995

2001

Canada United States Australia

1.24 2.34 0.95

1.60 2.72 1.311

1.72 2.51 1.662

1.94 2.82 1.533

Japan Korea

2.29 ..

2.93 1.92

2.89 2.50

3.09 2.96

Denmark Finland Norway Sweden

1.06 1.17 1.17 2.17

1.64 2.03 1.64 2.70

1.84 2.28 1.70 3.35

2.194 3.40 1.62 4.27

France Germany5 Italy Netherlands Spain

1.93 2.43 0.88 1.79 0.41

2.37 2.53 1.23 1.97 0.84

2.31 2.26 1.00 1.99 0.81

2.20 2.49 1.073 1.943 0.96

United Kingdom

2.38

2.07

1.95

1.90

EU OECD

1.69 1.95

1.90 2.23

1.80 2.10

1.93 2.33

1. 1990. 2. 1996. 3. 2000. 4. 1999. 5. Figures for Germany from 1991 onwards refer to unified Germany. Source: OECD.

© OECD 2004


150

Figure 5.4. R&D expenditure in manufacturing by technology intensity Percentage share in total manufacturing, 20001 High-technology

Medium-high technology

Medium-low- and low-technology

100

100

80

80

60

60

40

40

20

20

0

0 CAN

FIN IRE

UK USA

SWE KOR

FRA DNK

OECD ITA

EEC NLD

JPN BEL

NOR AUS

DEU ESP

CZE POL

1. 1998 for OECD and Norway, 1999 for Ireland, Denmark, France, Netherlands and EU. Source: OECD, ANBERD and STAN database, June 2003.

While overall R&D expenditure in the United Kingdom as a percentage of GDP is comparatively low, that expenditure is primarily in high technology sectors. The manufacturing sector in the United Kingdom, as in other countries, accounts for the largest share of private R&D expenditure and high-technology accounts for the vast majority of manufacturing R&D expenditure in the United Kingdom (Figure 5.4). Only Canada, Ireland, Finland and the United States have greater R&D expenditure by high-technology manufacturing industries. This composition of R&D expenditure, biased toward high-tech, may in part reflect a successful adjustment in manufacturing and mitigate the effects of the United Kingdom’s declining R&D expenditure overall arising from the falling share of manufacturing in the economy. Indeed, over the 1990s, high-technology industries increased their share of manufacturing R&D expenditure by 10 percentage points, a level markedly above the EU average (Table 5.5). Competition legislation and enforcement The overhaul of the United Kingdom’s competition policy framework is now virtually complete, but an unusually large number of institutions still apply competition law (Box 5.1). Their intricately related functions may provide some checks and balances, but potential overlaps may also create uncertainties and delays. Repeated rounds of reform have transformed the original 1947 law based

© OECD 2004


151

Table 5.5. Share of high-technology R&D spending in manufacturing % share in total manufacturing R&D, 1991-2000 1991

2000

Canada United States Australia

68.7 60.9 31.7

79.3 60.6 42.4

Japan

37.4

44.0

49.5 33.0 42.0 53.7

56.0 67.0 39.4 57.6

France1 Germany3 Italy Netherlands1 Spain

57.4 34.5 49.2 30.4 48.3

53.8 33.0 52.9 51.5 38.4

United Kingdom

50.7

60.2

Denmark Finland Norway2 Sweden

1

EU1 OECD2

46.9 52.4

1. 1999 instead of 2000. 2. 1998 instead of 2000. 3. 1995 instead of 1991. Source: OECD (2001a), Science, Technology and Industry Scoreboard.

on the “public interest” into competition-based rules that largely follow those of the European Union, except market reference provisions, which do not have an exact parallel at EU level. Much of this had been achieved in the Competition Act in 1998, which came into force in March 2000; and the recent Enterprise Act, which became effective on 20 June 2003 in respect of the competition and consumer law provisions, completes the fundamental changes. The Office of Fair Trading (OFT), and (in the case of markets) sector regulators with concurrent powers, may refer a merger, or a market which does not appear to be working competitively, to the Competition Commission (CC) for an in-depth investigation. Decisions in merger and market investigations at the CC now (since June 2003) depend only on competition-based tests. The CC now has powers to decide and implement remedies. While the government may still refer cases for investigation in exceptional circumstances on the basis of certain explicit public interest considerations, its ability to do so is tightly constrained by legislation. Among the host of changes in substantive standards and procedures, the most distinctive feature of the Enterprise Act is the introduction of criminal penalties for a cartel offence. After years of sustained attention to reforming the institutional structure, attention now must turn to how it is implemented. Some significant fines have been assessed, particularly for resale price maintenance. The OFT’s enforcement

© OECD 2004


152

Box 5.1.

Competition institutions

An unusually large number of institutions, most of them independent from the government, apply competition law in the United Kingdom. These include: – The Office of Fair Trading (OFT) is constituted as a board since 1 April 2003 and is now the principal decision-maker for non-merger matters, and decides most mergers at stage I. – The Competition Commission (CC), an independent body of experts has been a collegial investigative body since 1947 (although going under different names). Its responsibility has been to conduct investigations into mergers and monopolies, referred to it for in-depth investigation. It has had no role in applying the basic “antitrust” prohibitions. It has recently received new powers to make final decisions about remedies, independence from Ministers and a potentially broader scope for its “market” investigations. – The Serious Fraud Office (SFO) now has an important role in competition enforcement, as the prosecutor of the new criminal cartel offence. The OFT, however, can prosecute if the SFO does not. – The Competition Appeal Tribunal (CAT), the specialised court for competition matters, is now formally separate from the Competition Commission. Both a court and an expert body, the CAT has potentially important new powers to review decisions by the OFT and the CC in merger and market investigations and to hear and decide actions seeking damages due to competition violations. – The Department of Trade and Industry (DTI) no longer has a direct role in enforcement decisions to the extent they involve competition law and policy. Although the Secretary of State is trying to move into the background, DTI will continue to be a major player, with powers to make appointments, oversee performance, prepare legislation, and interpose considerations of the public interest. – Sectoral regulators for communications, transport, energy, and water have concurrent enforcement powers with OFT for Competition Act 1998 decisions and market references, but not mergers, subject to oversight by the CAT and the Competition Commission. The Concurrency Working Party among OFT and the regulators keeps communications open, avoiding potential overlap.

strategy has been to increase public awareness of the importance of competition and to establish a record of successful action. Targets of action or investigation have included independent schools, consumer-product brands such as Hasbro and cultural icons such as Manchester United. OFT’s important role in enforcing consumer protection regulation complements this attention to consumer issues in competition enforcement. The stronger law and penalties have been in place now for several years, yet there has been frustratingly little enforcement activity with regard to pure horizontal price fixing, although there are reportedly dozens of

© OECD 2004


153

cases being investigated. The ambitious market studies and investigations undertaken by the enforcement bodies, the OFT and the CC, may also have a significant effect on market conduct. These studies, which examine why particular markets may not be functioning well, have delved into areas ranging from banking and pharmacies to taxis and estate agents. In the area of merger control the Enterprise Act has adopted a “substantial lessening of competition” test. This standard is often considered to be better suited for economically sensitive analysis of oligopoly than the EU merger rule, which has until now been based on dominance.4 The United Kingdom does not require pre-notification, having concluded that the increase in enforcement effectiveness and the avoidance of costly post-merger remedies in a pre-notification system would not justify the burden on reporting businesses. The introduction of criminal penalties for hard-core cartels within the UK is also new. Targeting only horizontal practices, it is an offence for individuals to “dishonestly” agree with other persons that undertakings will engage in price-fixing, market-sharing, production or supply limitations, or bid-rigging. The law appears to create a per se rule, but it will have to be tested in court. Coverage and exemption issues There are few exemptions or exclusions to the competition law. However, special sectoral enforcement arrangements are unusually important in the United Kingdom. For example, while regulated monopoly sectors are not exempted from the law, it is typically enforced by the sectoral regulator. Sector – specific treatment remains for mergers in the water industry. Moreover, narrowly defined “public interest” considerations have been introduced, which can be invoked in merger cases involving national security, newspapers and other media. If these are invoked, then a merger is considered against both competition and the specified public interest grounds.5 The Enterprise Act has repealed provisions for exclusions for the self-regulatory rules of professional bodies. A handful of special exclusions and exceptions remain, and a consultation was launched in mid2003 about how to treat them. The most important element is a proposal to align the United Kingdom’s treatment of vertical restraints with the corresponding EU exemption regulation, which would simplify compliance for business. With the new competition framework, the government has stepped back from its historic role of deciding on remedies in merger and “monopoly” investigations. Decisions are now made by the CC, including decisions about remedies, and are to be determinative. Although in narrowly defined cases involving a public interest consideration the Secretary of State may reach a different decision based on public interest grounds from that which the CC has reached, this process will be transparent and is subject to rules. In general, the Secretary of State has the power to overrule any enforcement action in UK competition law on the basis

© OECD 2004

154


of “exceptional public interests” – but only if Parliament has, following a debate, approved the public interest consideration to be used. This power to introduce additional public interest considerations has never been used, although the power to do so remains in the statute book. Sanctions and enforcement The stated intention by the government to apply criminal penalties aggressively is unusual among competition enforcement agencies in Europe. The introduction of criminal penalties implies that the hardest-core cases could lead to jail time. Whether this threat will improve enforcement and deterrence against price-fixing remains to be seen. Criminal sanctions will only apply to conduct after the effective date of the law, that is, 20 June 2003. The Enterprise Act also introduced a nonpenal individual sanction. For any breach of the Competition Act prohibitions or Articles 81 and 82 of the EC Treaty, an individual can be disqualified by the court from serving as a company director for up to 15 years. The level of fines against enterprises, several of them in the range of £20 million, is unprecedented in the United Kingdom. Before the 1998 reforms, a finding of infringement would lead only to an order to desist. OFT cases involving large fines have included resale price maintenance of toys and of sports wear bearing team insignias, and pricing and marketing restraints on the distribution of a patented drug. While no substantial fine has been imposed on a purely horizontal price fixing scheme, the OFT contends that resale price fixing, its principal enforcement target so far, also dampens horizontal competition. The threat of large financial penalties is significant enough that the OFT’s leniency programme is producing results. So far, investigations prompted by leniency applications have led to three infringement decisions. Recent reforms cautiously supplement public enforcement with private initiatives. The OFT is committed to respond within 91 days to complaints regarding market problems (but not about the conduct of a particular firm) that are submitted by designated consumer organisations. Although the government supports empowering consumers in enforcement, it appears reluctant to give too much rein to private initiative. It is not clear that this caution is warranted. More private action could act as an important complement to the enforcement agencies. The Competition Appeal Tribunal (CAT) can now award damages after an infringement has been established. And, in a small step towards class actions, the CAT can decide representative claims for damages brought by specified bodies on behalf of groups of named individual consumers. Some private suits have also been brought before the ordinary courts, but mostly under EU competition law. Market studies and investigations Market studies and investigations enable the enforcement agencies to examine problems in markets that are more complex than simply an infringement

© OECD 2004


155

of competition law and can potentially have a great impact on competitive conditions in a market or the economy. Perspectives and powers of OFT studies differ somewhat from CC inquiries. OFT studies produce analysis and recommendations, and may often challenge policy makers to take action concerning regulations with anti – competitive effects. Although these are only advisory, the government has committed to responding to those recommendations within 90 days and publishing an action plan where appropriate. Because of OFT’s consumer enforcement role, an OFT study might examine an issue about information or redress that is not strictly a competition problem. An OFT study may lead to a CC reference when there are competition problems. The CC has more formal powers to gather information, and the CC’s investigations into problems in markets can produce tailored remedies. These “market” investigations, which succeed the “monopolies” investigations under the previous legislation, maintain the Commission’s historic role of investigating and making recommendations about market situations referred to it by Ministers. Under the previous legislation the Commission recommended what ought to be done if it found that conditions were not consistent with the public interest and it was up to the Minister to decide whether any remedy, such as regulation, was needed. Now, the Commission is concerned only with competition issues and not the public interest in general, unless there is an intervention by Ministers, in which case the CC will report on public interest issues, but its recommendations are not binding. In addition, it will be able to decide and implement remedies that do not depend on ministerial discretion. The Commission’s findings on competition tests and remedies are determinative, although the Secretary of State retains power over remedies for the limited number of cases involving specific “public interest” considerations. The Commission also has powers to make recommendations about remedies that it does not itself have power to implement, usually involving action by government or some other third party, and the government has undertaken to consider and respond to such recommendations within 90 days. The government’s response to OFT and CC recommendations will test its tolerance for independent advice and action. Regulatory policies While the United Kingdom has one of the least restrictive regulatory environments, this does not mean there is no scope for improvement. This section examines developments and outstanding problems in several service sectors, including inherently competitive service sectors such as retail distribution and the professions and a set of network industries where parts of the industry are inherently non-competitive and require regulatory oversight. In potentially competitive service industries such as retail distribution, the recent tightening of planning restrictions is most likely partially responsible for the continued gap in perfor-

© OECD 2004

156


mance relative to comparable OECD countries. However, in professional services the OFT has recently tackled the anti-competitive effects of regulations by professional bodies and recent reforms should greatly improve performance in the sector. In network industries, the United Kingdom was an early starter in opening up these industries to competition and in implementing many regulatory reforms and best practices that were then followed by other countries (e.g. RPIX price regulation). However, it took over a decade for effective competition to emerge in the telecommunications and energy sectors. In spite of the increased competition in these sectors, market power on the part of incumbent firms is still a concern. The effects of recent trends towards consolidation in the electricity sector will merit close scrutiny by the regulator. Reforms in the rail sector have been far from successful. The sector witnessed the failure of the infrastructure company, Railtrack, at the end of 2000 and has now been replaced by a “not-for-dividend” company. Retail distribution The distribution sector, which includes the wholesale and retail sectors, is subject to a host of regulatory restrictions. Boylaud and Nicoletti (2001) constructed an indicator of the restrictiveness of regulation in retail distribution in 1998. While overall the United Kingdom had one of the least restrictive regulatory environments in 1998, this was not the case in retail distribution, where the indicators show that the United Kingdom was close to the OECD average (Figure 5.5). While shop opening hours have been deregulated,6 the sector is still characterised by measures restricting market entry arising from planning restrictions. In general, while land use restrictions may be intended to achieve social objectives (e.g. preventing urban sprawl) they can have adverse effects by restricting efficient operation and thus lowering productivity and can protect incumbent firms from competition, thus creating or maintaining rents. The retail distribution sector benefits from economies of scale and scope and in many countries, including the United Kingdom, this is manifest in the ongoing process of structural change involving larger retail outlets, consolidation into retail chains and greater concentration and vertical integration. Considerable consolidation has occurred over the last decade and, compared with other OECD countries, the United Kingdom’s retail distribution sector is characterised by relatively high concentration. In food retailing, the combined market share of the five largest retailers is 63 per cent, considerably higher than the EU average of around 50 per cent (Figure 5.6).7 While, productivity growth in the sector compared with other G7 countries has been a relatively good during the past decade, partly reflecting the structural changes that have occurred, it has been half that achieved in the United States (Table 5.1).8 The sector is also characterised by one of the lowest outlet densities in the EU with an average number of employees per enterprise that is the highest in the EU and more than twice the EU average (Table 5.6).

© OECD 2004


158

Figure 5.6.

80

Five-firm market concentration in food retailing in EU countries

80

A. 1999

70

70

60

60

50

50

40

40

30

30

20

20

10

10

0

30

ITA

GRC

ESP

DEU

EU15 1 FRA

NLD

DNK

AUS

IRE BEL-LUX

UK

PRT

FIN

SWE

0

30

B. Changes in percentage points from 1993 to 1999

20

20

10

10

0

0

-10

-10

-20

-20

-30

ITA

GRC

ESP

DEU

EU15 1 FRA

NLD

DNK

AUS

IRE BEL-LUX

UK

PRT

FIN

SWE

-30

1. Weighted average. Source: Estimates based on data from Corporate Intelligence on Retailing’s European Retail Handbook. Reported in Dobson et al. (2001).

willing to pay a mark-up for their services (Dobson and Waterson, 1999). Yet, there are instances where the impact of out-of-town developments has had a major impact on town centres (e.g. the Meadowhall development near Sheffield). In response to these concerns the government significantly tightened planning policy in the mid-1990s in favour of town-centres and against out-of-own “retail parks”. The new planning policies have generally been successful in that since the mid-1990s the amount of new retail floor space in town centres has increased, contributing to the regeneration of town centres (Figure 5.7). However, the stricter land use and planning regulations have also made it difficult for large-format operators to develop new sites or expand existing ones.11 This limits the ability of large operators to compete and to achieve their full productivity potential and new operators are discouraged from entering the market.

© OECD 2004


157

Figure 5.5. Summary indicators of regulation in retail distribution 19981 5

5 Barriers to entry

Business restrictions

Prices

4

4

3

3

2

2

1

1

0

0 CZE

AUS CHE

CAN DEU

NLD KOR

IRE

SWE MEX ICE UK DNK TUR ITA GRC AUT HUN NOR ESP PRT FIN BEL POL JPN FRA

1. The scale of indicators is 0-6 from least to most restrictive. Source: Boylaud (2000).

The low outlet density and high number of employees per outlet might be expected to be conducive to high productivity. However, productivity or value added per person employed falls short of the levels in most comparable countries (e.g. France and Germany) and is around the EU average.9 A possible explanation for this is that compared with the United States and other large European countries, the United Kingdom still has a higher proportion of food retailing by smaller stores with a lower level of productivity (MGI, 1998). In addition, even the larger stores are small by international standards.10 In general, the structural changes that have occurred over the past decade have been beneficial for consumers. Larger retail outlets, providing one-stop shop services, offer greater convenience and lower prices arising from improved efficiency and resulting cost savings that are passed on to consumers. Regulatory easing in many countries has reinforced these structural changes, involving the closing down of scale – inefficient outlets and improvements in productivity. Large stores with modern formats are by far the most productive outlets in retailing and over time are likely to displace smaller traditional stores that are less productive. It is often claimed that there is a cost for consumers due to the decline of traditional retailing, offering specialisation or location convenience. In general, the impact of out-of-town developments in the United Kingdom on smaller traditional retailers located in town centres has not materialised (Fernie, 1998; MGI, 1998). Evidence seems to suggest that small shops still survive because consumers are

© OECD 2004


159

Key structural features of the retail distribution sector

Table 5.6.

2000 Outlet density1

Employees per enterprise

Value added per employed person2

Value added per unit of labour cost2

43 80 47 46 64 35 36 130 54 150 133 65 36 71 68 56

7.7 3.5 8.1 5.0 4.2 9.0 9.3 2.2 8.5 2.5 2.8 4.3 14.2 6.3 6.0 6.8

108 109 103 132 133 113 95 81 80 44 73 130 99 100 112 201

98 95 99 110 104 116 .. 72 117 81 97 88 123 100 98 ..

Austria Belgium Denmark Finland France Germany Ireland Italy Netherlands Portugal Spain Sweden United Kingdom European Union Norway Switzerland

1. Number of enterprises per 10 000 inhabitants. 2. EU = 100. Source: Eurostat, New Cronos.

Figure 5.7. New retail floor space in town centres and out of town New floor-space, gross letable area, thousand square metres, 1965-2000

800

800

600

600

400

400

200

200

0

0

200

200

400

400

600 800

600

Town center Non-town centre

800

Retail park

1000 1965

1970

1975

1980

Source: Experian, Oxford Institute of Retail Management.

© OECD 2004

1985

1990

1995

1000 2000


160

There are concerns that concentration and consolidation may lead to a lessening of competition. The scope for anti-competitive behaviour however is limited by the threat of entry and by increasingly mobile consumers. However, entry into retailing since the mid-1990s has not been that easy and, even where entry is possible, incumbent firms typically have a strategic advantage due to better location and reputation effects (Dobson and Waterson, 1999). Four out of the five major retailers have invested heavily in building their own retail brands and consumer loyalty through the use of loyalty cards. Own-labels represented almost 40 per cent of the UK grocery market in 1995, which is high compared with other European countries where the share represent anywhere from 7 to 20 per cent of the market (Table 5.7). Concerns have also been expressed about the buying power of retailers in the United Kingdom and its effects on economic welfare.12 Table 5.7. Own-label penetration in European packaged grocery Percentage of market

France West Germany Italy Spain UK Source:

1980

1990

1995

7 5 5 2 23

14 14 6 6 31

17 20 8 7 38

Dobson and Waterson (1999).

These developments have had an important economic impact in the distribution sector in the United Kingdom. Dobson et al. (1998) find that net margins in food retailing increased from 4 per cent in the early 1980s to 6.5 per cent by the mid-1990s and conclude that the higher margins may reflect the exercise of increased market power, either in squeezing suppliers or in raising prices to consumers. However, another possibility is that the increased margins during this period reflected the efficiency improvements that went hand in hand with the structural changes i.e. the development of large retail outlets and increased consolidation. Indeed, since the mid-1990s, and coinciding with the tightening of planning policies that have inhibited the further development of large retail outlets, margins have been falling (Table 5.8). Comparisons with overseas retailers show that UK firms have higher gross margins, possibly reflecting the much higher penetration of own-brand products which carry greater margins for retailers, but that operating margins are similar.13 A number of these issues were addressed in recent OFT and CC inquiries into supermarkets. The CC found that in general the UK market is competitive, and that price differentials between the United Kingdom and the EU could mainly

© OECD 2004


161

Table 5.8. Measures of profitability in food retailing1, 2 Per cent 1999 1996

1997

1998

1999

2000 UK

Gross margins Operating margins ROCE4

24.7 6.1 18.5

24.5 5.7 17.6

24.9 5.6 17.7

25.1 5.6 17.0

.. 4.5 13.3

25.1 5.0 15.4

Overseas

22.9 4.73 18.5

1. For the UK the measures are a weighted average of five major UK supermarkets (Asda, Morrison, Safeway, Sainsbury and Tesco) and for overseas firms they are a weighted average of Wal-Mart, Safeway Inc., Ahold, Delhaize, Carrefour, Métro and Promodès. 2. International comparisons are restated to reflect differences in depreciation. 3. Figures excluding Metro and Promodès. 4. Return on capital employed. Source: CC (2000).

be explained by exchange rates and other factors such as building and land costs. Some of the pricing practices of the major retailers reflected market power, and were found to be against the public interest. However, no remedies were recommended as it was felt that the remedies would require monitoring and a level of intervention that would be disproportionate to the adverse effects they were designed to fix.14 As regards practices with suppliers, the CC did find that large retailers could dictate trading relations with suppliers which distorted competition in the supplier market and was against the public interest. However, how protection against powerful, opportunistic buyers should be afforded to suppliers remains contentious.15 The CC recommended that the most effective way to remedy these adverse effects would be through a legally enforceable Code of Practice.16 The CC inquiry also examined issues concerned with planning and found that both entry into and expansion within one-stop shopping had become more difficult due to the change in planning guidelines in the mid-1990s. The Commission found no reason to suggest changes to the planning regime or the balance of interests with broader social and environmental objectives which the new planning guidelines seek to achieve. The CC noted, however, that the planning system was not designed to safeguard competition and consumer choice in multiple grocery retailing. There are trade-offs between the productivity improvements that come from larger retail outlets and the revitalisation of town centres arising from the new planning guidelines, and these trade-offs may be more difficult in the United Kingdom due to a high population density and overcrowding. Still, the current planning restrictions may be overly biased toward obtaining social objectives. They tend to insulate incumbent firms and limit innovation and competition in retail format strategies to the detriment of productivity improvements and cost savings that could then be passed on to consumers.

© OECD 2004


162

Professional services Professional services are usually subject to pervasive regulation, including the exclusive exercise of certain functions, entry and access requirements, recommended or fixed prices, and restrictions on advertising and business structure and nationality or residency requirements. Such intervention is often explained by the need to correct market failures, which are mostly due to information and transactions costs. However, this is not borne out by recent empirical evidence. Restrictions on competitive practices such as price competition and advertising or nationality requirements do not explicitly address the issue of quality, and there is little empirical evidence to suggest that these restrictions improve consumer welfare (Nguyen-Hong, 2000; OFT, 2001a; Paterson, et al., 2003). In practice these restrictions have been correlated with higher prices and less innovation, without improving quality and it appears that the principal beneficiary of such restrictions is the profession itself. Nguyen-Hong (2000) examined the effects of regulations on price-cost margins in engineering services and found that regulations led to an increase in prices on the order of 10 to 15 per cent in countries with the most restrictive practices. Paterson et al. (2003) found a negative correlation between productivity and the degree of regulation, and no evidence that less restrictive regulation led to a lower quality of services. The authors also found that countries with a high degree of regulation tend to have relatively higher turnover from fees, indicating higher mark-ups.17 These results support the view that restrictive regulatory frameworks and self-regulation by professional bodies, rather than supporting the needs and interests of consumers, are often used by the professions to obtain economic rents. In comparison with other EU countries the United Kingdom has a low or moderate degree of regulation in the accountancy, legal, architectural, engineering and pharmaceutical professions. The only other countries with such a profile are Ireland and the Netherlands (Table 5.9).18 In spite of such a low regulatory index, a moderate level of restrictions still exists in the legal and pharmaceutical sectors. In addition, competition in the professional services sector in the United Kingdom remains rather weak due to self-regulation by professional bodies. In light of these competition concerns the OFT recently conducted an investigation of the professional services sector in England and Wales (OFT, 2001a). The aim of the review, which focused on law, accountancy and architecture, was to identify restrictions which had the effect of preventing, restricting or distorting competition in the professions. The report recommended that: – Restrictions on price competition by fee guidance and, in particular, recommended fee scales published by the Royal Institute of British Architects and by the Law Society should be abolished. – Restrictions on advertising, particularly comparative fee advertising and direct marketing to potential clients by telephone (cold calling) should

© OECD 2004


Table 5.9.

Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom

163

Regulation indices for professional services1 Accountants

Legal

Architects

Engineers

Pharmacists

6.2 6.3 2.8 3.5 5.8 6.1 5.1 3.0 5.1 5.0 4.5 n.a. 3.4 3.3 3.0

7.3 4.6 3.0 0.3 6.6 6.5 9.5 4.5 6.4 6.6 3.9 5.7 6.5 2.4 4.0

5.1 3.9 0.0 1.4 3.1 4.5 n.a. 0.0 6.2 5.3 0.0 2.8 4.0 0.0 0.0

5.0 1.2 0.0 1.3 0.0 7.4 n.a. 0.0 6.4 5.3 1.5 n.a. 3.2 0.0 0.0

7.3 5.4 5.9 7.0 7.3 5.7 8.9 2.7 8.4 7.9 3.0 8.0 7.5 12.0 4.1

1. The higher the degree of regulation (intensity), the higher the respective figure (within a range from 0 to 12). All the regulation indices with a value of 5 or higher are shown in dark grey boxes, indices between 2.5 and 4.9 are in light grey boxes, and, those below 2.5 have a white background. 2. This figure only measures England and Wales. Source: Paterson et al. (2003).

be eased or removed. For example, barristers and accountants are prohibited from making direct fee comparison with other members of their profession and barristers are prohibited from referring to their success rate. – Indirect restrictions on entry including restricted consumer access (e.g. most clients cannot see a barrister without a solicitor) and restrictions on permitted business structures where professional rules and government regulation prevent multi-disciplinary partnerships (e.g. solicitors and barristers are prevented from practicing in partnership with one another and with other professionals such as accountants and financial advisers) should be lifted. – The entitlement to request that professional rules be excluded from the Competition Act prohibitions on anti-competitive agreements should be removed. – That the distinction between Queen’s Counsel (QC) and junior barristers be reviewed by the government as there are questions as regards the operation of the system as a mark of quality and its value to consumers. Some of these recommendations have been implemented (OFT, 2002). The entitlement to request exclusion form the Competition Act has been removed and professional bodies have undertaken a number of actions towards removing unjustified restrictions, or, in some cases, have justified the existing restrictions.19 In particular, within both the legal and the accounting profession, the restrictions

© OECD 2004

164


on comparative advertising of fees have been removed and there has been substantial easing of restrictions on direct marketing to potential clients by telephone.20 While much progress has been made, areas where serious restrictions to competition remain in place include continued restrictions on multi-disciplinary practises, especially in the legal profession. Legislation should permit a wide range of organisational solutions and ease the rigid rules on the incompatibility between professions. The government has launched a regulatory review of legal services which is expected to be completed by end 2004.21 The OFT also recently concluded an investigation into entry control regulation in retail pharmacy services (OFT, 2003). The OFT found that entry control regulations were unduly impeding the way that the market works and concluded that these regulations should be lifted. The OFT’s analysis concentrated on the effects on competition, the costs of administration, and the potential benefits of freedom of entry in terms of better service and lower prices. The report found that entry restrictions inhibited price competition, stifled efficiency improvement and innovation, and limited the availability of pharmacy services. In particular: – Since entry regulation was introduced in 1987, there has been very little entry into the sector. The regulations have acted to impede entry and expansion by pharmacies that offer lower prices or more convenient opening hours.22 – While the structure of the market has changed somewhat since 1987 (e.g. existing national chains and supermarkets have increased their market share), the effect of entry regulations has been to constrain such change and has impeded efficiency improvements. – The OFT estimated that deregulation could lead to substantial administrative cost savings on the order of £26 million and to annual customer savings on the order of £25-30 million.23 – Over the long term, improvements in efficiency could also be expected to reduce the overall cost of National Health Service (NHS) prescription drugs to the taxpayer. In responding to the OFT report the government raised concerns over rural and poorer areas and the current shortage of pharmacists. Although agreeing with the goals of lower prices, better service, and wider access, the government’s initial statement stressed the importance of a “balanced” package. However, the OFT considered that a shortage of pharmacists need not constrain entry over the longer term as new people enter the profession. Its report found that, by international comparisons, the United Kingdom has a slightly above average number of pharmacies per person and that consumers are currently well served geographically.24 The situation was complicated further by the fact that the OFT report appeared when the terms of the NHS contract were under review and that much of the power in this area has been devolved. Thus, the formal response, delivered as

© OECD 2004


165

promised within 90 days, preserves most of the status quo. The local bodies that control entry will be able to consider consumer choice and the benefits of increased competition in deciding whether a new or relocated pharmacy is necessary. Pharmacies in large shopping areas, at primary care centres, or opening much longer hours would be favoured.25 But the structure of entry control, based more on controlling competition than promoting it, remains in place. Network industries Network sectors in the United Kingdom (i.e. electricity, gas, water, transport and communications) account for about 12 per cent of value-added and 6.5 per cent of employment, and for a large share of intermediate inputs. Performance in these sectors is therefore important and can impact overall economic performance. There is now a solid body of cross-country evidence that liberalisation policies in network industries have led to higher productivity, better quality and, often, lower prices.26 However, capturing these benefits is not straightforward and close attention needs to be paid to the design of reforms (Gonenc et al., 2001). The United Kingdom has been at the forefront of many of these reforms, an example of which is RPIX price regulation which is now being applied by most OECD countries. A report by the National Audit Office (NAO) found that the way the regulators had applied the RPIX price controls in the United Kingdom had brought considerable benefits to both consumers and companies. Companies had become more efficient, and these efficiencies had not come at the expense of any reduction in quality of service (NAO, 2002). The sectoral regulators all share an essential remit to protect consumers against potential abuses of monopoly power, to promote or facilitate competition and to regulate prices where necessary. Competition legislation as regards anticompetitive agreements or abuses of a dominant position is typically enforced by the sectoral regulator whereas the OFT exercises competition powers with respect to mergers or market investigation. There are coordination agreements between the OFT and the sector-specific regulators with concurrent competition powers (OFT, 2001b). A Concurrency Working Party, chaired by the OFT with members from the sectoral regulators, meets on a regular basis and before the formal powers under the Competition Act are used, the OFT and the regulator agree which one of them will use the powers.27 The regulators have the power to impose financial penalties but must take account of the statutory guidance on penalties issued by the OFT. Funding of the regulators generally comes from a mixture of fees and charges levied on the industries they regulate. Telecommunications In the telecommunications sector, an independent regulator (Ofcom) is charged with promoting competition and the interests of consumers and with

© OECD 2004

166


regulating the sector, including interconnection, price regulation and universal service. Ofcom was created by the Communications Act 2003 bringing together the former telecoms regulator, Oftel, with four other regulators for the broadcasting industry. Regulation is primarily effected through the General Conditions of Entitlement under which all operators provide service. These replace the former licensing-based framework. The new regulator reflects issues of convergence between broadcasting and telecommunications, and is the UK’s means of implementing the provisions of a raft of new EU Directives modernising the regulatory framework for electronic communications which needed to be implemented by 25 July 2003. The regulatory regime in the United Kingdom has been characterised by structural measures that aim at improving competitive conditions. These include the use of long-run incremental costs (LRIC) and RPIX for determining regulated charges; carrier pre-selection (which has been available to British Telecom (BT) customers since 1986); number portability for both fixed and mobile numbers (implemented in the United Kingdom in advance of EU legislation); and recently, local loop unbundling (OECD, 2002b). While the United Kingdom’s regulatory regime in telecommunications is one of the most pro-competitive in OECD countries, competition has been slow to take off. In part this may be due to the fact that BT was privatised as a vertically integrated monopolist and that until the early 1990s the industry operated under a duopoly policy, with Mercury Communications being the only firm to compete with BT in the fixed network. It wasn’t until the early 1990s that competition began to take off with the opening up of the market. In 2001, new entrants in the United Kingdom had the highest share of access lines among OECD countries and among the highest market shares in national long-distance and international markets (OECD, 2003b). The market is very dynamic and labour productivity growth over the 1990s has been one of the highest amongst the G7 countries. Although the United Kingdom now has one of the most competitive telecoms markets in the OECD with new entrants having a higher share of the market relative to entrants in other countries, BT has still managed to maintain a dominant position in almost all markets (Figure 5.8). While the pro-competitive regulatory framework has resulted in important benefits for consumers in some areas (e.g. residential telephone and internet access charges are amongst the lowest in the EU), prices to business consumers are the highest in the EU and mobile telephony prices remain comparatively high (Figures 5.9 and 5.10). The relatively high mobile telephony prices reflect the fact that interconnection charges for call termination from fixed to mobile are amongst the highest in the European Union (EC, 2002b). A recent Competition Commission enquiry found that the mobile network operators (MNOs) have a monopoly on call termination on their own networks; that competitive pressures at the retail level did not constrain termination charges; and that charges were 30 to 40 per cent in excess of a “fair” charge (CC,

© OECD 2004


Figure 5.8.

167

Estimates of incumbent operators’ market share Retail revenue, 2001

Per cent

Per cent

120

120 Local calls Long-distance calls International calls Calls to mobile networks

100

100

80

80

60

60

40

40

20

20

0

0 BEL

DEU

GRC

ESP

FRA

IRE

ITA

NLD

SWE

UK

Source: EC (2002b), Eighth Report on the Implementation of the Telecommunications Regulatory Package.

2002).28 Oftel has also taken significant action to ensure competitive access to BT’s local loop via local loop unbundling (LLU).29 However, progress in rolling out LLU in the United Kingdom has been comparatively slow. In large part this may be due to the fact that the United Kingdom has some of the highest prices in the EU for both full unbundling and shared access (Table 5.10). Based on the disappointing results with regard to unbundling, Oftel will need to remain vigilant and ensure that tariffs are transparent and cost – oriented and that access is non-discriminatory. Energy sector Reforms in the electricity and gas sectors in the United Kingdom started earlier than in most countries (over a decade ago) and have gone further in introducing competition and structural separation and winding back price regulation. Achieving vertical separation has been crucial in facilitating the establishment of competition in these sectors (EC, 2002c). The potentially competitive segments of the industry contain a large number of private sector operators while the noncompetitive network component is privately held and regulated by an independent sector regulator, the Office of Gas and Electricity Markets, Ofgem. Reforms in this sector have contributed to the extremely rapid labour productivity growth rates over the 1990s (Table 5.1). Ofgem’s regulatory powers include the setting of

© OECD 2004


168

Figure 5.9. Average monthly telephone charges August 2002

Euros

Euros

60

60

A. Composite basket of residential telephone charges 1,2

50

50

Fixed Usage

40

40

30

30

20

20

10

10

0

0 SWE

UK

NLD

AUT

DNK

ESP

DEU

IRE

FRA

GRC

ITA

BEL

FIN

PRT

Euros

Euros

120

120

B. Composite basket of business telephone charges 1,3

100

100

Fixed Usage

80

80

60

60

40

40

20

20

0

0 SWE

DNK

NLD

AUT

ESP

GRC

IRE

FRA

FIN

BEL

ITA

DEU

PRT

UK

Euros

Euros

60

60

C. Basket of medium user mobile telephone charges 2

50

50

Fixed Usage

40

40

30

30

20

20

10

10

0

0 PRT

FIN

FRA

DNK

AUT

GRC

NLD

SWE

BEL

ESP

ITA

UK

DEU

IRE

1. Composite basket includes international calls and calls to mobile networks. 2. Including tax. 3. Excluding VAT. Source: OECD, Communications Outlook 2003.

© OECD 2004


169

Figure 5.10. Costs of internet access1 September 2002 Euros

Euros

140

140 Fixed public switched telecommunication network Public switched telecommunication network usage Internet service providers

120

120

100

100

80

80

60

60

40

40

20

20

0

FRA

DEU

UK

FIN

AUT

GRC

DNK

ITA

ESP

PRT

IRE

NLD

SWE

BEL

0

1. For 40 hours at day-time discounted Public switched telecommunication network rates, including VAT. Source: OECD (2003b), Communications Outlook, Paris.

Table 5.10. Prices for unbundled local loop EUR, 2002 Full unbundling

Shared access

Monthly rental

Connection

Monthly rental

Connection

Belgium Denmark Germany Greece Spain France Ireland Italy Luxembourg Netherlands Austria Portugal Finland Sweden United Kingdom

13.3 8.3 12.5 11.5 12.6 10.5 16.8 11.1 15.8 13.5 10.9 13.8 14.7 11.3 16.2

79.9 45.4 70.6 123.4 20.0 78.7 121.5 91.4 185.6 79.0 54.5 82.9 216.0 165.2 140.3

3.2 4.1 4.8 6.9 4.8 2.9 9.0 2.8 7.5 5.6 5.5 7.9 7.0 5.4 7.0

86.5 178.4 74.9 141.9 27.0 78.7 123.4 81.0 196.2 89.2 109.0 173.8 160.0 118.0 186.5

EU average

12.8

103.6

5.6

121.6

Source:

EC (2002b), Eighth Report on the Implementation of the Telecommunication Regulatory Package, Annex 5.A1.

© OECD 2004


170

network access tariffs, service quality standards and conditions governing access and interconnections to the network. It is funded through annual contributions paid by the service providers and has the highest staff levels and budget compared with other energy regulators in EU countries, raising the question as to why it takes such a large staff to regulate the market (Table 5.11).

Table 5.11.

Austria Belgium Denmark Finland France Germany Greece Ireland Italy Luxembourg Netherlands Portugal Spain Sweden United Kingdom

Competencies and resources of energy sector regulators Network access conditions

Dispute settlement

Staff number

Annual budget 2002 EUR million

R(elec)/R (gas) R/R R/R R/R R/R N/N M/n.a. R/R R/R M and R R/H R/n.a. M/M R/R R/R

R/R R/R R/R R/R R/n.a. C/C R/n.a. R/R R/R R/R C/C R/n.a. R/R R/R R/R

45 68 30 15 80 n.a. 43 31 86 2 55 52 153 33 330

9 15 3 1 9 n.a. 4 6 18 n.a. 6 7 19 3 58

Note: R – regulator responsible, M – ministry responsible, C – competition authority, N – not regulated, H – hybrid, n.a. – no regulator. Source: EC (2002c).

Electricity generation, transmission and distribution were split into vertically separate privatised companies in 1990, and gas transmission and distribution are also vertically separated. As a result of these early reforms, concentration in electricity generation and concentration in gas are the lowest in the European Union (Table 5.12 and Table 5.13). Competitive markets now exist in both gas and electricity wholesale market. The benefits from increased efficiency have been shared amongst consumers in terms of lower prices and improved and more innovative services. Energy prices, for both electricity and gas, are below the EU average with the exception of electricity for household consumers, where prices are slightly above the EU average (Table 5.14). The introduction of competition in both electricity and gas markets has led to a large number of consumers switching suppliers in order to capture the benefits associated with competition.30 As a result of increased retail competition, price controls in retail markets in gas and electricity were lifted in April 2002.

© OECD 2004

20021 Unbundling

Large users switching suppliers

Transmission2 Distribution2

Per cent

Generation

Retail sales

HHI3

20-30 2-5 n.a. n.a. 10-20 20-30 0 10-20 > 50 10-20 20-30 5-10 10-20 n.a. > 50

45 96 78 45 92 64 97 97 69 n.a. 59 82 83 90 36

67 53 38 33 90 50 100 90 72 100 48 99 94 47 42

2 028 6 118 4 018 2 472 9 606 1 756 10 000 9 418 5 560 8 158 1 814 4 008 2 466 2 538 1 044

Date of full opening of demand

Austria Belgium Denmark Finland France Germany Greece Ireland Italy6 Luxembourg Netherlands Portugal Spain Sweden United Kingdom

2001 2003 2003 1997 – 1999 – 2005 2004 – 2003 2003 2003 1998 1998

L L L O M L L L L M O L O O O

A L L M A A A M L A M A L L L

Concentration (CR3)

Concentration

Potential competition from imports4

21 25 39 22 12 11 12 7 14 100 19 30 4 29 3

Network access charge5

Medium voltage

Low voltage

20 15 15 15 15 25 15 10 10 20 10 15 15 10 n.a.

65 n.a. 25 35 50 55 n.a. 40 n.a. n.a. 35 n.a. 45 40 40


© OECD 2004

Table 5.12. Electricity market indicators and implementation of the EU Electricity Directive

1. Information for 2002, except for concentration measures which are for 2000. 2. Unbundling concerning operators. A = Accounting, L = Legal, M = Management and O = Ownership. 3. Hirshman-Herfindahl Index (HHI) in electricity generation. A market is generally considered competitive with a HHI< 1 000; moderately concentrated with 1 000 < HHI < 1 800; and highly concentrated with a HHI > 1 800. 4. Per cent of domestic generating capacity. 5. Estimated average charge, EUR/MWh. 6. The market opening in 2004 concerns non-household customers only. Source: EC (2002c), AEEG (2002).

171

Gas market indicators and implementation of the EU Gas Directive

172

Table 5.13.

20021 Unbundling

Date of full opening of demand

Transmission

Austria Belgium Denmark France Germany Ireland Italy Luxembourg Netherlands Spain Sweden United Kingdom

2002 2003 2004 – 2000 2005 2003 – 2003 2003 2006 1998

L L L A A M L A M O A O

2

Distribution

L L L A A M L A A L A O

2

Large users switching suppliers

Concentration of available gas

Per cent

Per cent

Oecd Economic Surveys: United Kingdom 2003-2004

OECD Economic Surveys: United Kingdom 2011

OECD Economic Surveys: United Kingdom - Volume 2007 Issue 17 (Oecd Economic Surveys) (Oecd Economic Surveys)

Oecd Economic Surveys: United Kingdom 2004-2005

OECD Economic Surveys: United Kingdom 2011

OECD Economic Surveys: Brazil (Oecd Economic Surveys)

Oecd Economic Surveys: Italy (Oecd Economic Surveys)

OECD Economic Surveys: United States 2010

OECD Economic Surveys: United States 2010

Oecd Economic Surveys: Mexico

OECD Economic Surveys Sweden 2007 (Oecd Economic Surveys) (Oecd Economic Surveys)

OECD Economic Surveys Ireland 2006 Issue 3 (OECD Economic Surveys)

Oecd Economic Surveys 2005: Japan (OECD Economic Surveys)

OECD Economic Surveys: Greece 2009

OECD Economic Surveys: Turkey, 2006

OECD Economic Surveys: Greece 2011

Oecd Economic Surveys 2005: Spain

OECD Economic Surveys Italy: 2011

OECD Economic Surveys: Indonesia 2010

Oecd Economic Surveys 2003 Austria

OECD Economic Surveys: Austria 2011

OECD Economic Surveys France 2009

OECD Economic Surveys: France 2011

OECD Economic Surveys: Estonia 2011

OECD Economic Surveys 2006: Luxembourg

OECD Economic Surveys: Belgium 2011

OECD Economic Surveys: Australia 2010

Oecd Economic Surveys: Brazil 2006

OECD Economic Surveys: Euro Area

OECD Economic Surveys: China 2005

OECD Economic Surveys 2011: India

Oecd Economic Surveys: United Kingdom 2003-2004

OECD Economic Surveys: United Kingdom 2011

OECD Economic Surveys: United Kingdom - Volume 2007 Issue 17 (Oecd Economic Surveys) (Oecd Economic Surveys)

Oecd Economic Surveys: United Kingdom 2004-2005

OECD Economic Surveys: United Kingdom 2011

OECD Economic Surveys: Brazil (Oecd Economic Surveys)

Oecd Economic Surveys: Italy (Oecd Economic Surveys)

OECD Economic Surveys: United States 2010

OECD Economic Surveys: United States 2010

Oecd Economic Surveys: Mexico

OECD Economic Surveys Sweden 2007 (Oecd Economic Surveys) (Oecd Economic Surveys)

OECD Economic Surveys Ireland 2006 Issue 3 (OECD Economic Surveys)

Oecd Economic Surveys 2005: Japan (OECD Economic Surveys)

OECD Economic Surveys: Greece 2009

OECD Economic Surveys: Turkey, 2006

OECD Economic Surveys: Greece 2011

Oecd Economic Surveys 2005: Spain

OECD Economic Surveys Italy: 2011

OECD Economic Surveys: Indonesia 2010

Oecd Economic Surveys 2003 Austria

OECD Economic Surveys: Austria 2011

OECD Economic Surveys France 2009

OECD Economic Surveys: France 2011

OECD Economic Surveys: Estonia 2011

OECD Economic Surveys 2006: Luxembourg

OECD Economic Surveys: Belgium 2011

OECD Economic Surveys: Australia 2010

Oecd Economic Surveys: Brazil 2006

OECD Economic Surveys: Euro Area

OECD Economic Surveys: China 2005

OECD Economic Surveys 2011: India

Recommend Documents